EduStack Academy

Search "digital marketing course" today and you'll get thousands of options within seconds , from ₹2,000 self-paced video bundles to ₹2 lakh executive programs, all claiming to make you "job-ready." Most of them won't. Here's how to actually tell the difference before you pay for one.

Start With the Question Most People Skip: What Does "Job-Ready" Actually Require in 2026?

Before evaluating any specific course, it's worth being honest about what employers are actually hiring for right now. Current Indian hiring data points consistently toward a specific combination: real proficiency in performance marketing (Google Ads and Meta Ads specifically), working analytics skills through GA4, genuine SEO fundamentals, and increasingly, practical fluency with AI marketing tools , not as a bonus skill, but integrated into how campaigns actually get planned and executed day to day.

A course that teaches these as disconnected modules, covered once each and never revisited, produces graduates who can define terms in an interview but can't actually run a campaign. A course built around live, hands-on work , where you're managing real budgets, reading real data, and adjusting real campaigns , produces something genuinely different: a portfolio and a working skill set, not just a certificate.

The Checklist: What a Genuinely Strong Course Should Include

Live campaigns, not just simulated exercises. The single clearest signal separating a course that builds real skill from one that doesn't: do you actually run live Google Ads and Meta Ads campaigns with real (even small) budgets, or do you only work through pre-built case studies and templates? Managing a live budget , watching it underperform, diagnosing why, and fixing it , teaches judgment that no case study can replicate.

Genuine analytics training, not a single GA4 walkthrough. Analytics proficiency is one of the most consistently cited high-value skills in current hiring data, but a single overview session doesn't build real fluency. Look for a course where you're regularly pulling and interpreting your own campaign data throughout the program, not just being shown someone else's dashboard once.

AI tool integration woven throughout, not bolted on as one module. Since AI fluency combined with strategic marketing skill is increasingly what commands premium salaries, a course that treats AI as a single add-on lecture rather than something used consistently across SEO, content, ad copy, and reporting work is teaching yesterday's syllabus with today's buzzword attached.

A structured path from fundamentals to specialization. The strongest career growth in digital marketing comes from genuine depth in one or two areas , performance marketing, SEO, or analytics , rather than staying a generalist indefinitely. A good course should build broad fundamentals first, then give you a real opportunity to specialize before you graduate, so you're not starting your specialization from scratch on the job.

Mentor access, not just recorded video content. Marketing judgment , knowing why a campaign underperformed, how to interpret a confusing data pattern, when to trust an AI-generated recommendation and when to override it , is genuinely difficult to learn from video alone. Direct access to an experienced mentor who can review your actual campaign decisions is a meaningfully different learning experience than passively watching lectures.

A real portfolio by the end, not just a certificate. When you're job-hunting, a certificate tells an employer you finished a course. A portfolio of real campaigns you planned, ran, and can explain in detail tells them you can actually do the job. Prioritize any course structure that leaves you with the latter.

Red Flags Worth Watching For

Courses that promise guaranteed high salaries with no mention of skill-building specifics. Genuine salary growth in this field comes from demonstrated skill and portfolio strength, not from a certificate alone , be skeptical of any program leaning harder on income promises than on curriculum substance.

Syllabi that haven't meaningfully changed in years. Digital marketing tools, platforms, and best practices shift constantly , a course teaching the same static syllabus it used two or three years ago is very likely teaching outdated platform mechanics, especially around AI tools and paid advertising interfaces, which have changed substantially even in just the past year.

No mention of AI tools anywhere in the curriculum. Given how central AI fluency has become to high-paying roles in 2026, a course that doesn't address AI-assisted marketing workflows at all is missing a skill set employers are now actively screening for.

Entirely pre-recorded content with no live interaction or feedback. Passive video consumption alone rarely builds the judgment needed to actually run a campaign , look for programs that include live sessions, real assignments with feedback, and some form of direct mentor interaction.

Questions Worth Asking Before You Enroll

A few direct, practical questions to ask any course provider before committing:

Will I run actual live campaigns with real budgets, or only case studies and templates?

How is AI tool usage integrated into the coursework, specifically?

What does the portfolio I'll have by the end actually look like?

Is there direct mentor access, and how frequently?

How recently was the curriculum updated, and what changed?

What specific platforms and tools will I get hands-on practice with , Google Ads, Meta Ads Manager, GA4, specific AI tools?

If a program can't answer these clearly and specifically, that's a meaningful signal about how seriously the curriculum has actually been built.

Why Practical, Live-Project Training Matters More Than Ever

The gap between "theoretical knowledge" and "job-ready skill" has arguably widened in 2026, not narrowed , precisely because AI tools have made it easier than ever to produce content and run basic campaign mechanics automatically. What's left as the genuine differentiator is judgment: knowing what to ask an AI tool for, how to evaluate whether its output is actually good, and how to read real performance data and adjust strategy accordingly. That kind of judgment is built through doing real work under real feedback, not through passively absorbing course material.

The Bigger Picture

Choosing a digital marketing course is genuinely one of the higher-leverage decisions you'll make if you're serious about this career path , the right program builds real, demonstrable skill and a portfolio that gets you hired; the wrong one leaves you with a certificate and a vocabulary of terms you can define but not yet apply. Evaluate any course against the practical checklist above before enrolling, and prioritize hands-on, mentor-guided, live-project learning over passive content consumption every time.

Every few months, someone declares digital marketing “oversaturated” or claims AI is about to make the entire field obsolete. Neither claim holds up against the actual hiring and salary data coming out of India in 2026 , if anything, the gap between demand and skilled supply is widening, not shrinking.

The Demand Numbers Are Genuinely Striking

India needs somewhere between one and two million digital marketers by 2026, according to industry salary reporting , and supply simply isn’t keeping pace with that demand. That shortage isn’t theoretical: 93% of Indian businesses increased their digital marketing budgets in the past year, spanning everything from local cafes building their first Instagram presence to large enterprises running six-figure performance marketing programs. India’s digital advertising market itself is projected to reach roughly ₹59,200-69,856 crore, growing at a pace that consistently outstrips how many properly trained professionals are entering the field each year.

This demand isn’t confined to agencies or tech companies either. Small and mid-sized businesses across nearly every sector , e-commerce, education, healthcare, finance, retail , now rely on digital marketers to drive website traffic, generate leads, and build brand awareness, meaning the job market has genuinely broadened well beyond the “digital marketing agency” stereotype most people still associate with the field.

What You Can Actually Expect to Earn

The salary data across multiple 2026 industry reports converges on a fairly consistent picture. Freshers entering the field typically start somewhere between ₹2.5 LPA and ₹6 LPA, depending on skills, certifications, and location, with monthly entry-level pay generally falling between ₹20,000 and ₹40,000. That’s a genuinely competitive starting point compared to many traditional career paths requiring a full degree , and digital marketing is explicitly a skill-first industry, where demonstrated ability and a strong portfolio often matter more to employers than formal academic credentials.

The growth trajectory from there is where the field gets genuinely interesting. Mid-level professionals with 3-5 years of experience and real specialization typically move into the ₹7-20 LPA range, and senior marketers , particularly those who’ve built deep expertise in performance marketing, SEO, or marketing automation , can reach ₹20-65 LPA. Specific roles show even wider bands: social media marketing careers alone span from around ₹1.8 LPA for freshers up to ₹32 LPA for director-level roles, reflecting just how much specialization and proven results shift earning potential within a single discipline.

One detail worth noting for anyone specifically weighing B2B or SaaS careers: Indian SaaS companies have emerged as some of the highest-paying digital marketing employers in 2026, in some cases surpassing e-commerce for mid-to-senior roles , a genuinely useful data point if you’re choosing which industries to target once you’ve built core skills.

The Skills That Actually Command Premium Salaries Right Now

Not every digital marketing skill pays equally, and the gap between “generalist” and “specialist” salaries has widened meaningfully in 2026. The skills most consistently cited across current salary and hiring reports include SEO, performance marketing (Google Ads and Meta Ads specifically), content marketing, data analysis and GA4 proficiency, AI-assisted marketing tools, email marketing, short-form video creation, and conversion rate optimization.

One specific combination stands out repeatedly in current hiring data: performance marketing (Google Ads plus Meta Ads) combined with genuine GA4 analytics fluency and hands-on AI tool proficiency. A marketer who can competently run all three together , planning campaigns, reading the data, and using AI tools to speed up execution without losing strategic judgment , can often independently manage a brand’s entire paid growth strategy, and that combination alone is frequently worth ₹15-25 LPA to a D2C or SaaS company willing to pay for someone who doesn’t need three separate specialists to cover the same ground.

What About AI Replacing Digital Marketers?

This is worth addressing directly, because it’s the single most common hesitation holding people back from starting a digital marketing course in 2026. The consistent finding across current industry analysis is that AI is not shrinking the digital marketing job market , it’s transforming it. AI is increasingly handling repetitive, mechanical tasks , first-draft copy, basic reporting, routine bid adjustments , while creating new, higher-level strategic roles focused on directing AI effectively, interpreting results, and making judgment calls AI genuinely can’t make on its own.

The practical implication: the marketers most at risk in this shift aren’t the ones using AI, they’re the ones who never developed real strategic and analytical skills to begin with, and who were essentially just executing repetitive tasks manually. A properly trained marketer who understands strategy, data, and how to direct AI tools effectively is, if anything, more valuable in 2026 than in previous years , not less.

What a Genuinely Good Digital Marketing Course Should Actually Include

Given how skill-dependent this field is, the course you choose matters enormously , and not every program on the market delivers the same real-world readiness. A strong digital marketing course should include hands-on work, live campaigns, and exposure to real tools, not just theory delivered through slides and quizzes. At minimum, a properly structured program should cover SEO, paid advertising across Google and Meta, content marketing, social media strategy, analytics through tools like GA4, and increasingly, practical AI tool integration , not as an afterthought module, but woven through every discipline the way it’s actually used in real campaigns today.

What to Actually Do If You’re Considering This Path

Don’t wait for the “right time” , the demand gap is current, not future. With India short somewhere between one and two million digital marketers relative to current business demand, this isn’t a market you need to time perfectly; the shortage already exists today.

Prioritize specialization over staying a generalist for too long. The salary data consistently shows the biggest jumps come from genuine depth in performance marketing, SEO, or analytics , not from spreading yourself thin across every discipline without real proficiency in any of them.

Choose a course based on hands-on project exposure, not just syllabus breadth. A course that has you running real campaigns, analyzing real data, and building an actual portfolio prepares you far better for hiring conversations than one built purely around theoretical modules.

Treat AI fluency as a core skill to build alongside the fundamentals, not a separate track. Since the highest-paying roles increasingly combine strategic marketing knowledge with genuine AI tool proficiency, look for a course that integrates AI tools throughout rather than treating them as a bonus add-on.

The Bigger Picture

The data is fairly unambiguous: digital marketing remains one of India’s strongest career opportunities heading into the back half of 2026, with genuine salary growth potential, broad demand across sectors, and a skills gap that shows no sign of closing on its own. The field has changed , AI, performance marketing sophistication, and analytics rigor have all raised the bar for what “skilled” actually means , but the underlying opportunity, for anyone willing to build real, demonstrable skills, is arguably stronger now than it’s ever been.

Ask most Indian D2C brands what WhatsApp is for, and you'll get the same answer: order confirmations, delivery updates, the occasional customer complaint. New data from 2026 says that framing is now badly out of date , and the brands still treating WhatsApp as a support channel are leaving a genuinely large amount of revenue on the table.

The Numbers Behind the Shift

GoKwik's WhatsApp Commerce Intelligence Report 2026 analyzed 26 billion messages sent by more than 1,800 Indian D2C brands across four quarters, and the findings are striking. WhatsApp is increasingly functioning as a genuine customer acquisition channel, not just a retention and post-purchase tool , during the October-December 2025 festive quarter, 83% of all WhatsApp-driven orders came from first-time buyers, not repeat customers being nudged to reorder. Brands actively using WhatsApp marketing tools recorded median gross merchandise value growth 2.25 times higher than brands that didn't. Fashion brands led the pack, with top performers converting at nearly 2.5 times the category average.

The scale underneath these numbers is genuinely enormous. India has somewhere between 500 and 535 million monthly active WhatsApp users, making it by a wide margin the platform's largest market globally. And Indian consumers don't just have accounts , they spend meaningfully more time on WhatsApp than users in virtually any other market, and have fully normalized making purchase decisions inside a chat thread, not just receiving updates about purchases made elsewhere.

Why WhatsApp Specifically Works So Well in India

The engagement numbers explain why this channel outperforms nearly everything else in a D2C marketer's toolkit. WhatsApp campaigns in India routinely achieve open rates in the 90-98% range, compared to roughly 20-25% for email , and a large share of messages get read within five minutes of delivery. Conversion figures follow the same pattern: personalized WhatsApp messaging can lift conversion ratios by up to 112% compared to other channels, with typical conversions running 3-5 times higher than email across multiple industry reports.

There's a specific reason this pattern is stronger in India than in most other markets: trust travels through WhatsApp here in a way it doesn't through anonymous marketplace listings or generic ad creative. A conversation feels personal, a marketplace listing doesn't , and for considered purchases especially, that difference in perceived trust directly drives the conversion gap.

Where This Is Working Best: Tier-2 and Tier-3 Cities

One of the more commercially significant patterns in this data is where WhatsApp commerce is converting hardest: tier-2 and tier-3 cities, markets that have historically been expensive and difficult for D2C brands to reach efficiently through traditional paid social or search. WhatsApp's combination of near-universal reach, low data requirements, and comfort with conversational commerce over polished storefronts gives it a genuine structural advantage in exactly the markets where India's next wave of digital consumer growth is concentrated.

What "Doing WhatsApp Commerce Properly" Actually Looks Like

The brands seeing the strongest results aren't simply blasting broadcast messages , they're building structured commerce systems inside the chat interface itself:

Native payment integration removes the biggest friction point. WhatsApp Pay and UPI integration inside chat let customers complete a purchase without ever leaving the conversation or being redirected to a separate website , a meaningful reduction in the drop-off that happens whenever a customer has to switch context mid-purchase.

AI handles discovery and routine questions; humans handle complexity. The most efficient setups use AI-driven catalog browsing, product recommendations, and FAQ handling for the majority of inbound volume , typically 70-85% , while routing complex queries or high-value orders to a human salesperson for the remaining share. This combination scales conversational commerce without requiring D2C brands to staff a fully manual chat operation as volume grows.

Interactive, form-based flows reduce back-and-forth for larger catalogs. Newer WhatsApp Business capabilities let brands build structured, app-like interactions within the chat itself , product selection, address input, delivery scheduling, and payment, all without the customer needing to type free-form messages back and forth. This matters especially for brands with large SKU counts, where an unstructured conversation would otherwise take too many messages to complete a purchase.

Post-purchase engagement drives real repeat revenue. Order tracking, delivery updates, and well-timed reorder nudges for consumable or replenishable products (skincare, supplements, FMCG) through WhatsApp have been shown to meaningfully lift repeat purchase rates , a large share of Indian consumers now actively expect delivery tracking through WhatsApp specifically, not email or SMS.

A Genuine Warning Worth Taking Seriously

There's an important measurement gap brands need to plan for: WhatsApp commerce revenue is frequently under-reported in standard analytics, because purchases often happen entirely in-app or through direct payment links that don't carry the UTM parameters marketers rely on to trace attribution elsewhere. If your reporting stack doesn't account for this, you may be systematically undervaluing WhatsApp's actual contribution to revenue , a real risk when deciding how to allocate budget across channels. Supplementing digital attribution with simple post-purchase surveys asking "how did you hear about us" is a practical, low-cost way to correct for this blind spot.

What to Actually Do About This

Stop treating WhatsApp purely as a support and logistics channel. If your current use is limited to order confirmations and delivery updates, you're using a fraction of what the channel is capable of driving in direct revenue.

Build a genuine acquisition strategy for WhatsApp, not just retention. With 83% of festive-quarter WhatsApp orders coming from first-time buyers in the GoKwik data, this channel is demonstrably capable of driving new customer acquisition, not just nurturing an existing base.

Prioritize WhatsApp commerce investment if you're targeting tier-2 and tier-3 markets. These are precisely the geographies where WhatsApp's conversion advantage over other digital channels appears strongest, and where the next wave of Indian D2C growth is concentrated.

Fix your attribution before concluding a channel "isn't working." Given how commonly WhatsApp revenue gets under-counted in standard analytics, verify your actual measurement setup before assuming this channel underperforms relative to what the broader industry data suggests it should be delivering.

Combine AI automation with genuine human escalation, not full automation alone. The brands scaling most successfully use AI for volume and speed while preserving a human touch for complex or high-value conversations , full automation without a human option tends to erode exactly the trust advantage that makes WhatsApp commerce work in the Indian market in the first place.

The Bigger Picture

WhatsApp commerce in India has moved well past early experimentation into genuine, measurable infrastructure for D2C growth , and the brands still treating it as an afterthought behind Instagram and Google Ads are increasingly out of step with where Indian consumers actually make purchase decisions. For any brand serious about growth in India specifically, WhatsApp deserves the same strategic investment, measurement rigor, and creative attention as any paid acquisition channel , not the leftover budget and thinking it's traditionally received.

Name changes are usually the least interesting part of any product update. This one is different: Google's rebrand of Merchant Center Next back to simply Google Merchant Center in July 2026 arrived bundled with genuine functional changes that directly affect how your products get discovered, how your Shopping ads get built, and how much manual feed work you'll need to do going forward.

The Rebrand Itself Is Cosmetic , Confirm That First

If you manage a Merchant Center account, the name change requires nothing from you directly. Your account, product data, campaigns, login process, and bookmarks all remain completely unaffected , this is purely a documentation and platform-labeling update rolling out gradually across Google's interface and help resources. Worth knowing simply so you're not confused when the branding shifts under you, but not something requiring action on its own.

The Part That Actually Matters: Product Studio and Feed Intelligence Scoring

The real substance sits in two connected updates that shipped alongside the rebrand. Product Studio is an AI-powered tool built directly into Merchant Center, using Gemini to rewrite product titles, generate lifestyle imagery, enrich product attributes, and run image split-testing , turning what used to be manual feed optimization work into a largely automated, AI-assisted process. For DTC brands managing large catalogs, this is a genuinely significant time-saver, provided you keep a human reviewing the output rather than trusting it blindly.

Feed Intelligence Scoring is the second, arguably more consequential piece. It directly affects how your products perform inside Performance Max: incomplete or low-quality product feed data can cause your listings to shift away from high-intent Shopping placements into lower-converting display or video inventory, reducing your overall ROAS. In plain terms, feed quality isn't just a housekeeping concern anymore , it's now a direct, scored input into which ad placements your products are even eligible to compete for.

Why This Connects Directly to Google's Broader AI Search Push

This update doesn't exist in isolation. It's part of the same wave of changes unveiled at Google Marketing Live 2026, where Google introduced AI-generated shopping summaries that explain to a searcher why a specific product is a good match for their query, and confirmed that Merchant Center feeds are increasingly the raw material Google's systems use to answer conversational, AI-Mode-style shopping queries directly , not just to populate traditional Shopping ad units.

That's a meaningful shift in what "good feed data" actually means. A feed built purely to satisfy Shopping ad requirements is no longer the full job; the same feed data is now feeding AI-generated product summaries and conversational shopping answers, meaning genuinely descriptive, accurate, and complete attributes matter more than they did when feeds were built purely for traditional ad placements.

What Else Shipped Around the Same Time

A few related updates worth knowing if you run a larger or agency-managed Merchant Center setup:

Automatic account linking. Starting around June 6, 2026, Google began automatically linking eligible Google Ads accounts with Merchant Center for advertisers not yet running product ads, with a "Connect Now" option for anyone who wants to initiate the connection earlier themselves. If you manage multiple accounts across clients, this is worth watching closely , an automatic link creates a real relationship between business assets, and getting clarity on account ownership and admin rights before an automatic link happens is simpler than untangling it afterward.

Merchant Center for Agencies. Available in the US and Canada since March 2026, this gives agencies a centralized dashboard to manage multiple client portfolios from a single interface, with early issue detection designed to catch problems before they trigger account suspensions.

Creative Content and video mapping. A newer feature lets brands map video content , sourced from social channels, websites, or Product Studio itself , directly to product listings, with AI-powered mapping from YouTube specifically designed to boost both SEO and brand presence across Google's properties.

Loyalty program integration. Retailers can now personalize member-only pricing and shipping benefits directly inside product listings, with a dedicated loyalty optimization goal in Google Ads to help budgets prioritize high-value, loyalty-enrolled shoppers specifically.

What to Actually Do About This

Audit your feed completeness now, treating it as a ranking factor, not just a compliance checklist. Since Feed Intelligence Scoring directly influences whether your products get high-intent Shopping placement or get pushed into lower-converting inventory, incomplete titles, missing attributes, or thin descriptions are now a measurable performance cost, not just an administrative gap.

Test Product Studio's AI-generated content, but review before publishing. The efficiency gain is real for large catalogs, but AI-generated titles and lifestyle imagery should go through the same human review process you'd apply to any other AI-assisted marketing content, especially given how much scrutiny AI-generated commercial content is facing across the industry this year.

Confirm account ownership and admin structure before automatic linking reaches your accounts. If you're an agency or manage multiple stakeholder relationships around a single Merchant Center account, get clarity on who controls what before Google links accounts automatically on your behalf.

Start writing product data for AI shopping summaries, not just Shopping ad units. Since your feed is increasingly the source material for AI-generated "why this product matches your query" summaries, genuinely descriptive, benefit-focused product content is worth the extra effort beyond what a bare-minimum feed requires.

The Bigger Picture

This update is a clear signal that product feed quality has moved from a background technical task to a genuine, scored competitive factor shaping both traditional Shopping placement and the newer AI-generated shopping experiences Google is building across Search. Brands treating their Merchant Center feed as a living, actively optimized asset , rather than a one-time setup task , are positioned to benefit from both the AI tooling and the placement advantages this update ties directly to feed quality.

Every performance marketer knows the routine: a big sale is coming, so you manually bump the daily budget, loosen your ROAS target, watch it closely for a few days, and then, if you remember , you manually revert everything once the sale ends. Google just built a tool that does that entire sequence automatically, on a schedule you set in advance, with no risk of forgetting to switch it back.

What Actually Launched

Google Ads Liaison Ginny Marvin announced three bidding and budgeting updates on June 15–16, 2026: an expansion of Smart Bidding Exploration, a new beta called Promotion Mode, and a separate bidding target optimization change taking effect August 17, 2026. Together, they represent the most significant reworking of Google's automated bidding infrastructure so far this year.

Promotion Mode: The Feature Worth Understanding First

Promotion Mode is a new beta available for Search and Performance Max campaigns that lets advertisers schedule a temporary, self-closing change to two levers at once across a defined date range: your ROAS tolerance and your daily budget. Set it up for a flash sale, a seasonal ramp, or a product launch, and the campaign automatically boosts spend and loosens efficiency requirements during that specific window , then reverts to your normal settings on its own once the window closes, with no manual cleanup required afterward.

It's genuinely compatible with campaign total budgets too , a budget format Google expanded to Search, Shopping, and Performance Max campaigns earlier in 2026 , meaning advertisers using a fixed total spend model for a promotional period can schedule a temporary performance push without breaking their overall budget constraints.

The Distinction Most Coverage Got Wrong

A common early misconception was treating Promotion Mode as simply a rebrand of Seasonality Adjustments, Google's existing tool for signaling an expected conversion rate spike. Marvin herself clarified the distinction directly in response to advertiser questions: Seasonality Adjustments tell Smart Bidding to expect a temporary change in conversion rate , your actual ROAS or CPA target itself never moves. Promotion Mode does the opposite: it directly adjusts your ROAS tolerance and injects extra budget, an actual lever change rather than a forecast signal.

Because they answer genuinely different questions, the two tools compose rather than compete. For a major sale, a sensible approach is running both simultaneously , signaling the expected conversion-rate lift through Seasonality Adjustments while separately scheduling a Promotion Mode budget and tolerance boost across the same date range.

An Important Constraint Worth Knowing Before You Schedule Anything

Once a promotion window is active, you cannot change your baseline ROAS target mid-window , only the tolerance you've allowed around it. That makes the planning decision genuinely consequential: how much tolerance to allow, which campaigns qualify, and what your budget ceiling should be all need to be decided before your peak window begins, not adjusted on the fly once the sale is live and budget is actively being spent. Getting that calibration wrong before a major event like a holiday sale means discovering the mistake in real time, during your highest-stakes sales period, with actual budget already committed.

Smart Bidding Exploration: The Quieter, Equally Significant Update

The second update expands Smart Bidding Exploration , a feature that lets Google's AI bid on search queries with plausible but historically unproven conversion records, beyond what your current ROAS target would normally allow it to pursue. As of June 15, 2026, it's globally available for Performance Max campaigns without a product feed, across all languages, with a separate beta opening for standard Shopping campaigns and Performance Max campaigns that do include product feeds.

The mechanic is straightforward: you set a ROAS tolerance as a percentage , say, 10% below your actual target , and Google's system stays within that band while testing genuinely new, unproven traffic categories. Google's own internal data reported campaigns using this feature saw meaningfully higher unique converting user counts on average, a real signal that the feature surfaces genuinely new customer segments rather than just reshuffling existing conversion volume. It's worth noting this feature generally requires unconstrained budgets and removed CPC bid caps to function as intended , a tightly capped campaign won't get the full benefit of the exploration behavior.

How These Connect to the August 17 Bidding Change

These two features arrived as part of the same announcement as the previously covered bidding target optimization change taking effect August 17, 2026 , the update that tightens how budget-limited Target CPA and Target ROAS campaigns perform relative to their stated targets. Understanding all three together matters: Promotion Mode and Smart Bidding Exploration give you legitimate, sanctioned ways to intentionally expand beyond your stated targets for defined reasons , a sale, a genuine growth push , while the August 17 change specifically closes the door on campaigns quietly, permanently outperforming a stale target with no deliberate strategy behind it. Google is effectively saying: if you want more aggressive performance, use these new explicit tools to ask for it, rather than relying on the system's old tendency to quietly exceed under-set targets by default.

What to Actually Do With This

Map your Q3 and Q4 calendar for genuine promotion windows now. Flash sales, seasonal ramps, and product launches are exactly what Promotion Mode is built for , and since tolerance settings lock once a window starts, the planning has to happen well before the event, not during it.

Request Promotion Mode beta access if you have any seasonal spend planned. Given the scheduling and revert mechanics, this is a meaningfully cleaner solution than ad-hoc manual budget edits that someone has to remember to reverse , a real, recurring operational risk for any team managing promotional campaigns manually.

Test Smart Bidding Exploration on Performance Max campaigns with genuinely unconstrained budgets. Set a conservative tolerance , 10 to 15% , and let it run for a full two weeks before evaluating whether the new converting query categories are actually valuable for your business, not just technically "new."

Don't set Promotion Mode tolerance carelessly. Since the baseline ROAS target is locked once the window opens, treat the tolerance-setting decision with the same care you'd give any other irreversible-for-the-duration campaign setting , get it right in planning, not in the middle of your highest-stakes sales period.

The Bigger Picture

This release fits a consistent pattern across Google Ads in 2026: the platform is building more explicit, deliberate controls for advertisers who want aggressive performance during specific windows, while simultaneously tightening default behavior for campaigns that were quietly benefiting from loose, unmanaged targets. The advertisers who come out ahead are the ones treating these as genuine planning tools , mapping their promotional calendar and tolerance decisions in advance , rather than the ones discovering how any of these levers actually behave for the first time during a live, high-stakes campaign.

For years, B2B marketers have faced an uncomfortable trade-off: LinkedIn’s own ad platform has the professional targeting data that actually matters, job title, seniority, company, industry, but it comes at a premium CPC that makes testing expensive. Microsoft Advertising just closed that gap in a way no other search platform can replicate, because it owns the data source directly.

What Actually Launched

Microsoft Advertising added job seniority as a targeting dimension inside its LinkedIn Profile targeting feature on June 15–16, 2026, according to Product Liaison Navah Hopkins. The update lets advertisers target or observe audiences using 10 standardized seniority levels pulled directly from LinkedIn member profiles: CXO, VP, Director, Manager, Senior, Entry, Owner, Partner, Training, and Volunteer. It’s available across both Search and Audience campaigns, applicable at either the campaign or ad group level, and rolled out live in 29 markets spanning the Americas, EMEA, and Asia-Pacific, including the US, UK, India, Australia, and Japan.

This isn’t Microsoft’s first foray into LinkedIn-powered targeting. The capability traces back to April 2022, when Microsoft first introduced job function, industry, and company targeting drawing from LinkedIn’s professional network data. What’s new this time is specifically the seniority layer, a dimension that lets advertisers distinguish between a decision-maker and a practitioner within the exact same company and job function, something the platform previously couldn’t do.

Why This Is a Genuinely Structural Advantage, Not Just a Feature Update

Here’s the detail worth understanding clearly: Microsoft owns LinkedIn. That ownership means Microsoft Advertising is the only search ad platform in the world with access to declared LinkedIn profile data, not inferred professional attributes pieced together from browsing behavior, but the actual company, industry, job function, and seniority a real person entered on their own LinkedIn profile. Applied directly to Bing search intent, that combination is difficult for any competing search platform to replicate without acquiring LinkedIn itself.

The practical upshot for B2B marketers is significant: you can reach the same verified decision-maker audience LinkedIn’s own ad platform targets, but layered onto search intent instead of feed placement, typically at a meaningfully lower cost per click than LinkedIn’s native ad costs.

How the Targeting Actually Works

A few mechanical details matter for anyone planning to use this properly:

It’s a bid modifier, not a strict filter, mostly. On Search, Shopping, and DSA campaigns, LinkedIn profile criteria function as bid adjustments rather than hard delivery filters, meaning non-matching users can still see your ad; you’re adjusting how aggressively you bid for each seniority tier, not excluding others outright. Hard filtering and exclusion exist only on Audience (native/MSAN) campaigns specifically.

Each tier supports its own independent bid adjustment. Rather than a simple include-or-exclude toggle, advertisers can raise bids specifically for CXO or VP-level audiences while holding standard bids for Manager or Senior tiers, letting you incrementally weight spend toward higher-authority buyers without abandoning broader reach entirely.

Observation mode lets you gather data before you commit. You can layer seniority in purely as a reporting dimension, watching how each tier actually performs, without narrowing your campaign’s delivery at all. This is the sensible default for the first several weeks: gather real conversion signal by seniority tier before deciding where to concentrate bid increases.

The tiers aren’t equal in buying power, and treating them as interchangeable is a mistake. CXO, VP, and Director levels are the obvious high-authority targets for enterprise sales motions, but depending on your product and deal size, Manager or even Senior-level practitioners are sometimes the actual champions who drive a purchase decision internally, don’t assume seniority alone predicts influence for every business model.

What to Actually Do With This

Start in observation mode across all 10 tiers for at least two to four weeks. Resist the urge to immediately restrict targeting to CXO and VP alone, the data on which tiers actually convert for your specific offer is worth gathering before you narrow anything.

Build separate ad groups with tailored messaging by seniority band. An executive-facing ad group can lean into strategic outcomes and business growth framing, while a practitioner-facing ad group performs better emphasizing workflow efficiency and implementation details, the same product benefit, framed differently for who’s actually reading it.

Layer seniority bid adjustments on top of your existing job function and industry targeting, rather than treating it as a standalone campaign structure. Combined, these LinkedIn-sourced dimensions let you build genuinely precise account-based targeting inside a search campaign, something that historically required LinkedIn’s own, pricier ad platform.

Compare cost-per-lead directly against your LinkedIn native campaigns for the same audience segment. Since Microsoft is offering comparable declared-data targeting at typically lower CPCs, running a genuine side-by-side test against your existing LinkedIn spend is the clearest way to find out how much budget you could reasonably reallocate.

The Bigger Picture

This update reinforces Microsoft Advertising’s position as a genuinely distinct B2B targeting stack in 2026, not a cheaper alternative to Google Ads, but a platform with a specific structural advantage no competitor can copy without owning LinkedIn outright. For B2B marketers who’ve been paying LinkedIn’s premium CPCs purely for its professional targeting precision, this update is a legitimate reason to test whether a meaningful share of that budget performs just as well, or better, running through Microsoft’s search inventory instead.

Every marketing team claims to be “using AI” in some form these days. Very few actually show their board and investors the numbers to prove it changed anything. Urban Company just did exactly that, on its Q1 FY27 earnings call this week, and the specifics are a genuinely useful case study for any marketing leader wondering whether AI adoption is producing real results or just faster busywork.

The Numbers That Started the Conversation

Urban Company CEO and co-founder Abhiraj Singh Bhal told investors that the company’s India Consumer Services marketing spend was essentially flat year over year, ₹24 crore in the same period last year versus ₹25 crore this year, while the return on that spend improved meaningfully. “So actually, more or less flat marketing, but we’ve been getting a lot better ROI for our spends there,” Bhal said. That’s a genuinely rare claim to make on an earnings call with real numbers attached: not “we grew marketing efficiency,” but a specific, checkable statement that spend stayed nearly flat while returns improved.

What “AI-Powered Marketing” Actually Meant in Practice

Here’s the part worth reading carefully, because it’s more specific than the usual vague “we use AI” corporate line: Bhal described the entire marketing team as leveraging AI end-to-end, covering creative creation, campaign deployment, optimization of those campaigns, and learnings and redeployment based on what worked. That’s not a single AI tool bolted onto one part of the workflow, it’s AI touching every stage of the marketing loop, from the first draft of a creative asset through to feeding results back into the next campaign’s decisions.

This lines up precisely with the distinction researchers have been drawing all year between organizations that see genuine AI-driven business impact and organizations that don’t. Recent global research on this exact question found that while roughly 90% of companies report AI transforming their workflows, only about 18% see significant revenue impact from it, and the gap consistently comes down to whether AI gets embedded across an entire workflow and feeds back into decision-making, versus being layered on as an isolated tool for one task. Urban Company’s description, AI touching creative, deployment, optimization, and learning as one connected loop, is a textbook example of the pattern associated with actually seeing revenue results, not just the pattern associated with adoption alone.

It Wasn’t Limited to Marketing

Urban Company’s AI usage extends well beyond the marketing function, which matters for understanding why the company was willing to make such a specific public claim. The company also uses AI for customer and service-partner support, partner onboarding and training, quality audits, and fraud detection. Perhaps the most striking figure in the entire disclosure: Bhal said around 90% to 95% of Urban Company’s code is now being written using AI, creating real leverage in engineering costs and headcount.

The company’s India Consumer Services segment saw its Adjusted EBITDA margin expand by 170 basis points to 6.9% of NTV, which Urban Company attributed roughly equally to two factors: higher contribution margins driven mainly by AI-led support savings, and operating leverage from revenue growing faster than overhead costs. Notably, Urban Company did not attempt to precisely quantify how much of the marketing ROI improvement specifically came from AI versus other factors, a reasonably honest acknowledgment that isolating AI’s exact contribution inside a broader efficiency story is genuinely difficult, even for a company confident enough to discuss it publicly.

Why This Case Study Matters More Than the Usual AI Marketing Anecdote

Most public claims about “AI transforming our marketing” come from vendors selling AI tools, or from case studies without real financial context attached. This one is different in a specific, useful way: it’s a public company disclosing genuinely checkable numbers, flat spend, improved margin, a specific description of where AI sits in the workflow, on an earnings call where overstating results carries real reputational and regulatory risk. That doesn’t make every detail perfectly precise, but it does make this a meaningfully more credible data point than the typical marketing AI success story.

What This Means for Your Own Marketing Team

A few practical takeaways worth applying, based on what actually distinguished this example:

Embed AI across the full campaign lifecycle, not just one stage. Urban Company’s description explicitly covered creative, deployment, optimization, and learning as one connected process. Teams using AI only for first-draft copywriting, while running deployment and optimization manually, are capturing a fraction of the efficiency this kind of end-to-end integration can produce.

Build a genuine feedback loop from campaign results back into future decisions. The “learnings and redeployment” piece of Urban Company’s description is arguably the most important and most commonly skipped step, many teams use AI to produce content or manage bids, but few systematically feed campaign results back into how the AI approaches the next campaign.

Track spend and ROI with enough precision to actually make this kind of claim yourself. Urban Company could make a specific, credible statement about flat spend and improved returns because it had the underlying reporting discipline to measure both precisely. That level of measurement rigor is a prerequisite for proving AI’s value, not an afterthought.

Be honest about what you can and can’t isolate. Urban Company’s own restraint, not overclaiming exactly how much of its ROI improvement came from AI specifically, versus other factors, is a reasonable model for how to talk about AI impact credibly, rather than attributing every efficiency gain to AI regardless of the actual evidence.

The Bigger Picture

This is a genuinely useful real-world counterpoint to the widespread skepticism about whether AI-driven marketing efficiency claims hold up under scrutiny. Urban Company’s willingness to disclose specific numbers on a public earnings call, in a setting where investors and analysts can and will ask follow-up questions, lends real weight to the broader argument that AI, used properly and embedded across an entire workflow rather than bolted onto one piece of it, can produce measurable marketing efficiency gains, not just faster busywork.

Beating earnings estimates usually sends a stock up. On July 30, 2026, Reddit beat them convincingly, revenue climbed 61% year-over-year to $804.9 million, and its shares still fell more than 20%. The reason wasn’t the numbers. It was what CEO Steve Huffman said about Google right after delivering them.

What Actually Happened

Huffman used Reddit’s Q2 2026 earnings call, and a letter to investors published alongside it, to directly challenge Google’s AI Overviews feature, the AI-generated summary box that now sits above traditional search results for most queries. His central argument was blunt: “The 10 blue links have driven a tremendous amount of value for the whole ecosystem, and AI Overviews has yet to make a similar level of positive impact.” He described search referral traffic as having turned “choppy,” and said Reddit is “still just looking for that win-win” that the traditional search results format used to reliably provide.

The criticism landed with more weight than the usual publisher complaint about AI search, for one specific reason: Reddit is reportedly working to end its licensing agreement with Google, according to earlier Wall Street Journal reporting, a deal previously valued at around $60 million annually. The same report named several other major publishers, including The Economist, Reuters, Politico, and USA Today, as considering similar moves away from their own Google arrangements.

The Argument Underneath the Numbers

Huffman’s letter to investors went further than a straightforward traffic complaint, it made a philosophical case for why Reddit specifically shouldn’t be reducible to an AI summary in the first place. “As the internet becomes flooded with synthetic content, people are craving real human perspective,” he wrote, framing Reddit as “the antidote to an automated web.” His most quotable line captured the core argument directly: “AI compresses the internet into summaries. Reddit delivers the opposite: deep discussions, passionate debates, and lived experiences... People don’t want a summary of Reddit; they want Reddit.”

It’s a genuinely interesting position for Reddit to take, given how much the platform has benefited from AI in other respects this year, Reddit content has become one of the most heavily cited sources across AI search generally, and Reddit’s own visibility inside Google’s traditional organic rankings has been climbing sharply through 2026’s core updates. Huffman isn’t arguing against AI’s use of Reddit content broadly; he’s specifically arguing that a compressed summary destroys the exact thing that makes Reddit’s content valuable in the first place, the discussion itself, not just the conclusion someone might extract from it.

The Structural Problem Publishers Can’t Currently Solve

Buried in the coverage of Huffman’s comments is a detail that explains why this tension has been building for over a year without resolution: publishers currently cannot separate being included in Google’s traditional search index from having their content used inside AI Overviews. It’s a single, bundled package, opt into regular search visibility, and your content becomes fair game for AI summarization too, with no way to accept one and decline the other.

This is precisely the problem the new Search Console AI Overviews opt-out setting was built to solve, following a binding order from the UK’s Competition and Markets Authority. But that control remains in limited rollout, and Reddit’s frustration suggests the underlying tension it’s meant to address hasn’t cooled off in the meantime, if anything, the loudest, most Google-dependent publishers are getting more vocal about it, not less.

What Reddit Is Actually Doing About It

Notably, Huffman wasn’t just complaining, he outlined Reddit’s countermeasures directly. Rather than waiting for Google to fix the traffic economics, Reddit is focused on “converting some of the web traffic into direct traffic, where it becomes more valuable for us”, building product features and user retention specifically designed to reduce Reddit’s dependence on search referrals altogether. That’s a notable strategic pivot: instead of fighting to preserve the old click-through model, Reddit is trying to make search traffic matter less to its business model in the first place.

Why This Matters Beyond Reddit and Google

This dispute is a genuinely useful case study for any brand or content business still building its strategy around traditional search referral traffic. If a platform as large, well-resourced, and search-traffic-dependent as Reddit is publicly signaling frustration and actively working to reduce its reliance on Google search, going so far as reportedly negotiating to end a paid licensing deal over it, that’s a strong signal about where the wider industry’s leverage and priorities are shifting.

A few practical takeaways worth applying to your own content strategy:

Direct and owned traffic is becoming more valuable precisely because referral traffic is becoming less reliable. Reddit’s own stated strategy, converting web visitors into direct users, mirrors advice that applies broadly: email lists, app usage, and repeat direct visits are assets that don’t erode every time a search platform changes its algorithm.

Watch how the AI Overviews opt-out setting evolves, since bigger players pushing for it may accelerate its rollout. If major publishers keep escalating pressure the way Reddit just did publicly, genuine separation between traditional indexing and AI summarization may arrive faster than Google’s current gradual rollout timeline suggests.

Don’t assume your content’s presence inside an AI Overview is automatically valuable. Huffman’s argument, that a compressed summary can actually destroy the specific value of certain content types, is worth applying to your own content honestly. Some content genuinely loses its value when reduced to a three-sentence AI summary; other content survives that compression just fine. Knowing which category yours falls into should shape how urgently you push for direct control over AI summarization.

The Bigger Picture

A platform beating earnings expectations and still losing 20% of its market value in a single day is a strong signal that investors are taking the AI-search traffic disruption seriously, not as a temporary inconvenience, but as a genuine, unresolved structural risk to any business built on search referrals. Reddit’s public pushback this week may be the loudest version of this argument yet, but it’s very unlikely to be the last, and brands watching how this plays out have a real interest in how it resolves.

There’s a quiet win a lot of PPC managers have gotten comfortable relying on: set a Target CPA of $10(≈ ₹952), and over a few months, Smart Bidding settles the account at $5 instead, extra efficiency the algorithm found on its own, and you got to keep. Starting August 17, 2026, that quiet win disappears, and if you don’t act before then, your cost per acquisition could climb right back up to whatever number you originally typed in, whether or not that number still makes sense.

What’s Actually Changing

Google announced the change on June 15, 2026, through Ads Product Liaison Ginny Marvin, then followed up with additional clarification on June 22 after the PPC community raised significant pushback and confusion about exactly what the update would do. The core mechanic: starting August 17, 2026, campaigns that are “limited by budget” and running Target CPA or Target ROAS bidding will begin performing much more precisely toward the actual number you’ve entered as your target, not the better number Smart Bidding may have quietly discovered was achievable.

Previously, budget-limited campaigns benefited from an extra layer of optimization that most advertisers never had to think about: Google’s system would naturally gravitate toward the cheapest, highest-quality conversion opportunities available, often meaningfully outperforming the stated target as a side effect. After August 17, that hidden optimization goes away. If your campaign’s Target CPA is set at $10 but your actual recent performance has been running at $5, the campaign will start delivering results much closer to that original $10 figure, a real, potentially unwelcome increase in cost per acquisition for accounts that never bothered to update an old target after performance improved.

Which Campaigns Are Actually Affected

The change specifically applies to campaigns that are both limited by budget and using target-based bidding strategies. It covers Search, Shopping, Performance Max, Demand Gen, and Travel campaign types. App, Video reach, and Video view campaigns keep their current behavior, and Display and Hotel campaigns are already operating under this tighter bidding logic, so they aren’t newly affected. If your campaign isn’t budget-constrained, this change doesn’t directly touch it at all.

Why Google Is Making This Change

Google’s stated reasoning is about predictability: the company wants advertisers to get more consistent, predictable performance in line with the targets they’ve actually set, especially as campaigns scale or budgets get adjusted. From Google’s perspective, a target that’s being routinely and significantly beaten isn’t really functioning as intended, it suggests the number itself is stale, not that the account is running unusually well.

There’s a reasonable case for that framing. But the PPC community’s sharper counterpoint is worth taking seriously too: a target set deliberately above your true acquisition goal was never sloppy management, for many advertisers, it was a genuine strategy, giving Smart Bidding room to explore, test new audience segments, and surface conversions that came in cheaper over time as the algorithm learned. This update closes off that exploration room unless advertisers actively compensate for it.

What Won’t Happen Automatically

Google has been explicit and repeated on this point across its own documentation and public statements: your budgets will not increase automatically as a result of this change, and your CPA or ROAS targets will not be automatically adjusted either. The change affects purely how the algorithm optimizes toward your existing target, not the target number itself. If you want your bidding to reflect your account’s actual recent performance rather than a stale target, you have to make that update yourself, manually, before the rollout takes hold.

The Tool Google Built to Help

Starting July 6, 2026, Google began sending account notifications to any advertiser with a campaign that was limited by budget at any point in the last 12 months on an affected bidding strategy, directing them to a new Bid Target Adjustment Tool. The tool lets you review recent campaign performance and quickly apply updated targets that better reflect what your account has actually been achieving, rather than an outdated number nobody revisited.

What to Actually Do Before August 17

Pull a list of every campaign marked “Limited by budget” using Target CPA or Target ROAS. This is the specific, identifiable group facing real risk from this change, everything else is unaffected.

Compare your actual recent performance against your stated target for each one. If your real CPA or ROAS has been meaningfully better than the target for a sustained period, that’s the exact pattern this change is designed to correct, and the exact pattern that will cost you if left alone.

Use the Bid Target Adjustment Tool to update targets that no longer reflect reality. Rather than letting the rollout silently push your account back toward an outdated number, set your target deliberately based on what you actually want to pay, informed by recent performance data.

Treat this as a legitimate audit opportunity, not just defensive maintenance. August 17 is a clean, forcing deadline to review whether any of your targets were set somewhat arbitrarily months or years ago and simply never revisited as campaigns matured, a useful excuse to align your stated goals with your actual current business economics.

Pay particular attention if you’re a lead-generation or local service business. For categories like home services, where a modest increase in cost per lead has an outsized effect on margins, this change deserves genuine attention before the rollout, not after you notice costs creeping up in September.

Time your review with Q4 in mind. Since this rollout lands right as back-to-school and pre-holiday campaign spending typically ramps up, a sudden, unmanaged performance reset at exactly this moment is a worse outcome than the same reset happening during a slower period.

The Bigger Pattern

This update fits a broader theme running through Google Ads changes in 2026: the platform is steadily tightening the relationship between what advertisers explicitly set and what the algorithm actually delivers, reducing the space for “hidden” optimization that advertisers benefited from without fully understanding. Combined with the DSA-to-AI-Max migration and AI Max’s expanded default status, the throughline is consistent, Google wants advertisers making more deliberate, explicit choices about their bidding strategy, not relying on quiet algorithmic generosity to cover for outdated settings.

August 2, 2026 marks one of the most consequential regulatory deadlines marketing teams have faced in years. Article 50 of the EU AI Act, the provision requiring disclosure whenever AI is used to generate or manipulate content people see, becomes legally enforceable across the European Union, and the penalties for getting it wrong are large enough to demand real attention, not a quick policy update.

What Actually Takes Effect Today

The European Commission adopted formal guidelines on Article 50 just two weeks ago, on July 20, 2026, clarifying exactly how these transparency obligations apply. The rules require providers and deployers of AI systems to be transparent about AI use across four specific areas: direct interaction between people and AI systems (like chatbots), AI-generated or AI-manipulated content such as images, audio, video, or text, emotion recognition and biometric categorization tools, and, the one most relevant to marketing specifically, deepfakes and AI-generated text on matters of public interest.

The stakes are real. Non-compliance can attract fines of up to €15 million or 3% of a company’s worldwide annual turnover, whichever is higher, a penalty structure serious enough that this can’t be treated as a minor compliance footnote.

Why This Applies to Far More Businesses Than You’d Expect

A common misconception is that Article 50 only applies to companies building AI models. It doesn’t. The obligations apply to any organization that uses generative AI to produce or alter content, meaning marketing content, social media posts, websites, advertising, product descriptions, and audiovisual material all fall within scope wherever generative AI touched them, regardless of whether your company builds AI tools or simply uses ones built by someone else.

It’s also not limited to businesses physically based in the EU. If your marketing function generates synthetic content for campaigns targeting EU audiences, or if a customer-facing chatbot serves EU customers, you’re in scope, location of your headquarters doesn’t exempt you.

The Part That Should Concern Every Marketing Team Specifically

Here’s the detail worth reading twice: AI-generated images, audio, video, or text must carry a clear, machine-readable label identifying them as AI-generated, unless the content has been meaningfully edited by a human. That carve-out sounds generous until you look at how narrowly it’s actually being interpreted. Routine touch-ups to an image are exempt, but altering a product to look better than it actually is likely is not, meaning a fairly normal marketing practice, enhancing product photography, could trigger a disclosure obligation depending on how substantially the AI changed the underlying representation.

Deepfakes specifically, AI-generated or manipulated content depicting real, identifiable people, carry a stricter, more direct labeling duty under Article 50(4), separate from the general content-marking requirement.

There’s a Grace Period, But It’s Narrower Than It Sounds

Some relief exists: under the AI Omnibus provisional agreement from May 2026, generative AI systems already on the market before August 2, 2026 get until December 2, 2026 to meet the specific machine-readable marking requirement under Article 50(2). That’s a genuine four-month buffer for the technical marking mechanism, but it’s a mechanism deadline, not a free pass on the underlying transparency obligation itself. Deepfakes generated before today aren’t subject to mandatory retroactive labeling, though the Commission is encouraging it voluntarily.

The Voluntary Code That’s Become the Practical Compliance Path

Alongside the binding guidelines, the Commission published a Code of Practice on Transparency of AI-Generated Content on June 10, 2026, developed with independent experts through the European AI Office. It’s technically voluntary, but the Commission has been unusually blunt about the distinction, noting plainly that even though signing the code is optional, the transparency requirements under Article 50 remain legal obligations regardless. Signing the code is increasingly viewed as the recognized, practical route to demonstrating compliance, and roughly 190 companies and organizations had signed by the end of July, a useful early benchmark for what responsible implementation looks like in practice, even for businesses not ready to sign formally themselves.

What Marketing Teams Need to Do This Week

Inventory where AI touches your content pipeline. Website copy, social posts, product photography, video, audio, and any customer-facing chatbot all need to be mapped against whether generative AI was used to create or meaningfully alter them.

Determine who counts as the “deployer” in your organization. Compliance responsibility sits with whoever puts the AI system into use, for many marketing teams, that’s the marketing function itself, not just IT or a vendor, and getting this ownership question answered clearly now avoids confusion during an actual compliance review later.

Build disclosure into the creative workflow, not as an afterthought bolted on at the end. The businesses treating this well are integrating provenance and disclosure directly into how content gets produced and approved, not adding a disclaimer after a campaign has already shipped.

Review vendor and agency contracts. If outside agencies or AI vendors are producing content on your behalf, contractual language requiring their compliance with Article 50 marking and disclosure obligations is a sensible protection to add now.

Consider signing the Code of Practice, or at minimum benchmarking against it. Even without signing, using the Code’s technical and operational guidance as a reference point for your own labeling and marking practices gives you a defensible, documented approach if regulators come asking.

Don’t assume this is an EU-only problem if you’re a global brand. Any campaign targeting EU audiences, regardless of where your company is based, falls in scope, and building your labeling practices to the EU standard globally is often simpler than maintaining two separate content pipelines.

The Bigger Picture

This is one of the most concrete, binding tests yet of how seriously regulators are willing to enforce AI transparency in advertising and content, and it’s arriving at the same moment consumer research is independently showing that undisclosed AI use is already eroding brand trust on its own, without any regulatory pressure at all. Businesses treating today’s deadline as a genuine operational shift, rather than a legal technicality to quietly work around, are the ones positioned to build durable trust with EU audiences rather than face escalating scrutiny as enforcement matures.

For twenty years, “impressions” has been the currency every media plan gets built around. That’s starting to change, and the shift became explicit this past month when a room full of CMOs and media publishers at India’s ET Brand World Summit 2026 in Mumbai reached a notable point of agreement: raw reach and gross impressions no longer mean what they used to.

Why the Old Numbers Stopped Making Sense

The reasoning behind the shift is straightforward once you say it out loud: algorithmic feed consumption has made impressions increasingly meaningless as a standalone metric. An impression only tells you a piece of content was technically served to a screen, it says nothing about whether anyone actually looked at it, how long they engaged, or whether it left any impression on their memory at all. In a media environment dominated by infinite, algorithmically-served scrolling feeds, that gap between “served” and “seen” has widened to the point where the old metric barely describes reality anymore.

The agreement among media buyers reflected at the summit points toward a specific replacement framework: attention metrics. Instead of counting how many times an ad was technically displayed, marketers are increasingly measuring dwell time (how long someone actually looked at a piece of content), audio-on viewability (whether video content was watched with sound engaged, a strong signal of genuine attention rather than passive scrolling past muted autoplay), and downstream brand recall (whether the exposure actually left a lasting impression that shows up later, rather than evaporating the moment the next post loads).

Why This Matters Beyond One Summit

This isn’t an isolated regional trend, it reflects a reckoning that’s been building across the industry for a while, as marketers have grown increasingly aware that a technically-served impression and genuine audience attention are two very different things, and that budgets optimized purely for the former have been quietly wasting spend on content nobody actually processed.

The framing that’s emerged from this shift is a useful one for any marketing team to internalize: creative resonance and genuine engagement have effectively replaced raw ad impressions as the priority metric worth optimizing for. That’s a meaningfully different design brief than the one most media planning has run on for the past two decades, it asks not “how many people can we technically reach” but “how many people will actually stop, watch, and remember.”

What This Actually Changes About How You Should Plan Campaigns

If attention, not exposure, is becoming the real currency, a few practical shifts follow directly:

Creative quality becomes a media efficiency lever, not just a brand consideration. If dwell time and audio-on viewability are the metrics that matter, a mediocre ad that technically reaches a million people is worth less than a genuinely compelling one that holds attention for a fraction of that reach. Production investment increasingly pays for itself in measurement terms, not just brand-perception terms.

Sound-on design stops being optional. With audio-on viewability emerging as a real attention signal, content built to work only with sound off, captions doing all the work, visuals carrying the entire message, is measurably weaker under this framework than content designed to genuinely earn a viewer’s attention with sound engaged.

Brand recall needs to become a tracked metric, not an assumed outcome. Downstream brand recall requires actual measurement, surveys, brand lift studies, or platform-provided recall metrics, rather than simply assuming that reach eventually converts to memory. Marketing teams that haven’t built recall tracking into their reporting are flying blind on exactly the metric this shift says matters most.

Reporting to stakeholders needs to evolve alongside the metrics themselves. A media plan built around impressions and CPM is easy to explain to a finance team but increasingly disconnected from what actually drives business outcomes. Making the case for attention-based metrics to budget holders who are used to simpler, older numbers is itself a real communication challenge marketing leaders need to get ahead of.

The Broader Pattern This Fits Into

This shift connects directly to several other trends reshaping marketing measurement throughout 2026: the rise of zero-click search meaning fewer traditional traffic metrics are even available to measure in the first place, growing skepticism toward AI-generated and heavily produced content, and a general audience fatigue with anything that reads as manufactured rather than genuinely attention-worthy. Across search, social, and now traditional media buying, the consistent throughline is the same: raw volume metrics are losing credibility, and genuine engagement, attention, and trust are becoming the metrics that actually predict business results.

What to Actually Do About This Shift

Start by auditing your current reporting stack: how much of what you report to stakeholders is impression and reach data, versus genuine attention and recall data? If it’s heavily weighted toward the former, that’s a gap worth closing before your next planning cycle, not after. Where your platforms offer dwell time, watch-through, or recall-lift measurement, start incorporating it into regular reporting now, even informally, so you have a baseline before attention metrics become the industry standard everyone’s expected to report against.

The Bigger Takeaway

Impressions aren’t disappearing as a metric entirely, they still have a place in understanding raw reach. But the consensus forming among marketing leaders is clear: impressions alone no longer tell you whether your marketing actually worked. Brands that get ahead of this shift, building genuine attention and recall measurement into their strategy now, will be far better positioned to prove real business impact than brands still reporting on numbers that increasingly convince no one, including themselves.

If you’re running Dynamic Search Ads and breathed a sigh of relief when Google pushed the mandatory migration deadline back five months, take a breath, then get to work anyway. The extension isn’t a reason to wait. It’s the exact opposite.

What Actually Changed

Google announced on June 11, 2026 that the automatic migration of Dynamic Search Ads (DSA) to AI Max for Search, Google’s newer, more automated Search campaign structure, has been postponed from September 2026 to February 2027. Google Ads Liaison Ginny Marvin confirmed the change publicly the following day, explaining that Google heard advertiser feedback loud and clear about needing more time, and specifically wanted to avoid forcing a disruptive account change during the busy Q4 shopping season.

It’s worth being precise about what actually moved and what didn’t, because Google split this into two separate timelines running in parallel:

September 2026 still applies to: Automatically Created Assets (ACA) and campaign-level broad match settings, which remain on their original schedule and will transition to AI Max as planned.

February 2027 now applies to: the full automatic migration of any remaining active DSA campaigns and ad groups. Between now and then, advertisers can continue creating new DSA campaigns and use voluntary manual upgrade tools with full oversight over the process. That creation window closes in January 2027, immediately before the February automigration begins, a one-month gap specifically designed to stop last-minute new DSA campaigns from being created right before the forced transition.

For context on how generous this extension actually is: the original DSA retirement was announced on April 15, 2026, with roughly 5.5 months of runway to the initial September deadline. The extension now provides roughly 10 months from that original announcement, a notably longer transition window than Google gave advertisers during the 2022 Smart Shopping to Performance Max migration, which allowed about 9 months.

Why “More Time” Is Actually the Risky Part

Here’s the trap worth naming directly: an extended deadline tends to produce exactly the behavior it’s meant to prevent, procrastination. Advertisers who read “extended to February 2027” as “this isn’t urgent yet” are the ones most likely to end up scrambling in January, running an unplanned, unaudited, fully automatic migration instead of a deliberate, tested one.

The more useful way to read this extension is as a runway to run a real, controlled experiment before you’re forced into the new system. Google itself is explicitly encouraging advertisers to audit existing DSA campaigns, run side-by-side experiments comparing DSA performance against AI Max for Search, and use the voluntary migration tools well before the deadline, specifically because manual, deliberate upgrades preserve historical reporting and campaign learnings that get lost in an automatic, forced transition.

The Part That Should Actually Worry You

AI Max for Search doesn’t just replicate DSA with a new name, it genuinely expands how much control you hand over. Dynamic Search Ads already let Google choose which queries to match based on your site content rather than a fixed keyword list. AI Max widens that further, using AI-powered matching that reaches well beyond your existing keyword list entirely. Without deliberate governance, a real negative keyword strategy, close monitoring of the search terms report, and clear budget guardrails, that expanded matching can quietly find its way into query territory you’d never have consciously targeted, and your account can start spending on searches you never actually approved.

That’s the real argument for using this extended window productively rather than defensively: the accounts that come through this migration cleanly will be the ones that spent the runway building negative keyword governance and testing controls, not the ones that simply waited for the automigration date to arrive.

What Google Is Also Signaling About Its Broader Direction

This isn’t an isolated change. AI Max for Search is now enabled by default when creating new Search campaigns entirely, advertisers can still disable it, but the default has shifted. Google has reported observing faster first conversions during testing, particularly within the first two weeks after a new campaign launches under AI Max. Combined with the DSA migration, keyword targeting moving toward keywordless matching, and manual assets shifting toward Automatically Created Assets, the pattern across Google Ads this year is consistent: manual control surfaces are steadily being folded into AI-managed ones across the entire platform, not just in this one campaign type.

What to Actually Do With This Extra Time

Audit your current DSA campaigns now, not in January. Know exactly which campaigns, ad groups, and budgets are currently running on DSA before you start testing anything against them.

Run a genuine side-by-side test. Set up a real comparison between your existing DSA performance and an AI Max for Search campaign covering similar territory, and give it enough time and budget to produce a meaningful read before committing further.

Build your negative keyword and query monitoring strategy before you need it. Since AI Max’s matching genuinely extends beyond DSA’s already-broad approach, having governance in place before migration, not scrambling to add it after spend has already leaked, is the single highest-leverage use of this extended timeline.

Use the manual upgrade tools deliberately, rather than waiting for the forced transition. A planned, tested migration preserves your historical reporting and campaign learnings; an automatic one in February 2027 does not offer the same continuity.

Mark both deadlines on your actual calendar, not just the headline one. September 2026 for Automatically Created Assets and broad match, and February 2027 for the full DSA migration, accounts running multiple affected campaign types need a plan that addresses both dates specifically, not just the more publicized one.

The Bigger Takeaway

Google’s decision to extend this deadline is a genuine accommodation to advertiser feedback, but it changes the calendar, not the destination, every DSA campaign still ends up on AI Max eventually. The advertisers who come out ahead won’t be the ones who felt relieved by the extension; they’ll be the ones who treated the extra months as exactly what Google intended: time to test, govern, and migrate on their own terms, rather than Google’s.

For as long as Search Console has existed, it answered exactly one question: how does your website perform in Google Search? If a TikTok video or Instagram Reel of yours started ranking for a real search query, that performance simply vanished into a blind spot, measurable inside the platform’s own dashboard, but completely invisible to the tool marketers actually use to understand Google visibility. That gap just closed.

What Actually Launched

Google introduced platform properties on July 7, 2026, and confirmed on July 29 that the feature is now globally available to everyone. It’s a new Search Console property type, sitting alongside the familiar domain and URL properties, that lets you verify an Instagram, TikTok, X, or YouTube account directly and see exactly how that content performs inside Google Search, Google Discover, and Google News, including clicks, impressions, and the specific search queries that surfaced your posts.

This matters more than it might sound at first. Social and video content has been driving real, measurable search discovery for years, a well-optimized TikTok caption or a strong YouTube title genuinely can rank for a product or informational query, but until now, that performance was completely disconnected from the analytics tools SEOs and content marketers rely on daily. Platform properties finally puts that data in the same place, and in the same familiar format, as your website’s Search Console data.

What You Actually Get Access To

Each platform property comes with three reports, mirroring the structure marketers already know from standard Search Console properties:

Performance report. Total clicks, impressions, and the specific search queries and posts driving your traffic, filterable and sortable, and exportable if you’d rather analyze it in another tool.

Insights report. A higher-level view of recent traffic trends, your top-performing posts, and how people are actually discovering your account through Google.

Achievements report. Notable milestones tracked over a rolling 28-day window.

One meaningful limitation worth knowing before you get too excited: there’s currently no combined, cross-platform view. Each platform property reports independently, meaning you’ll need to check your Instagram, TikTok, X, and YouTube performance separately rather than seeing everything in one unified dashboard. Building a combined view right now requires manually exporting and comparing data yourself, or connecting a third-party dashboard tool to the Search Console API for each property individually.

Who This Is Actually Built For

Google has been explicit that platform properties is aimed specifically at creators and brands who don’t run a website of their own, not just traditional publishers layering on a new data source. That’s a genuinely notable inclusion, for the first time, someone building an audience purely through TikTok or YouTube content, with no owned domain at all, gets structured, first-party Google Search data about their own visibility. Previously, that kind of insight simply didn’t exist inside any Google tool.

For brands and marketers who do run a website, the real value is comparative: you can now see, in one consistent format, how your owned content and your social content each perform in Google’s ecosystem, and start making genuinely informed decisions about where to invest production effort based on real search-discovery data, not just each platform’s own internal engagement metrics.

What You Can Actually Do With This Data

Google’s own guidance points to a few concrete, practical uses worth building into your workflow:

Find content worth resurfacing. Use the Performance report to identify older videos or posts that are unexpectedly regaining search traction, a strong signal that content is worth pinning to your profile, refreshing, or expanding into a follow-up piece.

Track the impact of edits over time. Search Console annotations let you mark when you make an external change, rewriting a YouTube title, updating a TikTok caption, and then watch whether that change actually affects your search performance, turning what used to be guesswork into something closer to a real experiment.

Compare formats directly. Use page filters or comparison mode to evaluate how different content types perform against each other, short-form versus full-length video, for instance, or one video playlist against another, using real Google Search traffic data rather than each platform’s internal engagement numbers alone.

What to Actually Do This Week

Verify every platform account you actively manage. If you or your brand publishes regularly on Instagram, TikTok, X, or YouTube, claiming the corresponding platform property costs nothing and immediately starts building a data history you’ll want later, the sooner you connect, the more historical comparison data you’ll have to work with down the line.

Start comparing your social content’s search performance against your website’s. If a particular topic performs dramatically better as a YouTube video showing up in Search than as a blog post on your own site, or vice versa, that’s genuinely actionable intelligence about where to invest your next piece of content, not just a curiosity.

Watch for query overlap and gaps between platforms. Since each property reports independently for now, manually comparing which search queries surface your website content versus your social content can reveal genuine content gaps, topics where you’re winning on one channel but completely invisible on another.

The Bigger Picture

Platform properties is part of a broader pattern this year: Google is increasingly treating the full, fragmented landscape of where content actually lives, not just traditional websites, as part of its measurement and discovery ecosystem. For marketers, that’s a genuinely useful development. The blind spot between “how my content performs on the platform itself” and “how my content performs in Google Search” has quietly cost marketers real strategic insight for years, and closing it gives content teams a much clearer, first-party picture of where their actual search visibility is coming from.

Google’s new Search Console setting lets publishers exclude their content from AI Overviews and AI Mode without losing regular search rankings. Here’s exactly how it works, what you gain, what you lose, and how to decide.

Quick Summary

For the first time, publishers can choose to keep their content out of Google’s AI Overviews, AI Mode, and AI Overviews in Discover, while staying fully indexed and ranked in regular search results. The control began rolling out in the UK on June 3, 2026 under a binding order from the UK’s Competition and Markets Authority, took effect June 17, and is now expanding to a broader set of site owners. It’s a genuine strategic decision, not a simple toggle, this guide walks through exactly what it controls, what it doesn’t, and how to decide if it’s right for your site.

Why This Setting Exists

Until this control launched, publishers faced a blunt, all-or-nothing choice: let Googlebot crawl your site and accept that your content could appear anywhere Google decided to place it, including inside AI-generated summaries that answer a searcher’s question without ever sending them to your page, or block Googlebot entirely and disappear from Google Search altogether. There was no middle ground.

The UK’s Competition and Markets Authority changed that with a legally binding conduct requirement, the world’s first mandate of its kind, ordering Google to give publishers genuine opt-out controls over AI Overviews and AI Mode specifically. Google complied by adding a dedicated setting inside Search Console, under Settings → Search generative AI.

What the Setting Actually Controls

It’s important to understand precisely what this toggle does and doesn’t affect, since the scope is narrower than many site owners initially assume.

Covered by the opt-out:

AI Overviews (the AI-generated summary box at the top of regular search results)

AI Mode (Google’s fully conversational search experience)

AI Overviews inside Google Discover

Not covered by the opt-out:

Standard organic search rankings and indexing, completely unaffected

Google Merchant Center participation

Google Ads eligibility

The Gemini app specifically (a separate product from AI Overviews and AI Mode)

Google has confirmed that opting out does not result in any ranking penalty elsewhere in Search. A site that opts out is not downranked in regular results, it simply won’t be summarized, linked, or cited inside the three AI features the setting covers. Google can still use the pages to help its systems interpret language and queries generally, even from opted-out sites.

The New Wrinkle: Top Stories Carousels Inside AI Overviews

This decision just got more complicated for news and content publishers specifically. Google confirmed on July 17, 2026 that a Top Stories news carousel is now fully live inside AI Overviews for developing-topic queries, currently limited to US mobile users, with UK figures showing the carousel appearing inside the AI Overview roughly 17.5% of the time Top Stories shows up at all.

Here’s the tension this creates: Google’s own documentation states that opted-out sites will not be linked within AI Overviews or the other covered features, and since the Top Stories carousel is a set of publisher links displayed inside the AI Overview, industry analysts reading Google’s documentation closely believe opting out likely means losing that carousel placement too. Google has not explicitly confirmed this specific interaction, but it’s considered a high-confidence read based on how the feature is documented, since Google’s opt-out rule applies to the whole AI Overview feature, not selectively to parts within it.

For news publishers specifically, that raises the stakes considerably: opting out now potentially means forfeiting not just general AI Overview citations, but also a fast-growing distribution channel for breaking and developing news content.

How to Decide: A Simple Framework

There’s no universally correct answer here, the right choice depends entirely on your own traffic data and business model. Work through these questions honestly before deciding:

1. Is AI Overviews summarizing your content and satisfying the search without sending you a click? If your analytics show strong AI Overview appearances but declining referral traffic, that’s the clearest signal opting out could protect real value you’re currently losing for free.

2. Or is Google’s AI Overview citing you and sending you qualified visitors? If AI-referred traffic is showing up in your analytics and converting reasonably well, staying in captures a genuinely growing discovery channel, Google puts AI Overviews above 2.5 billion monthly users and AI Mode above 1 billion, a scale too large to walk away from lightly.

3. Does your business model depend on visibility, or on direct monetized traffic? A site funded by ad impressions or affiliate clicks has a much stronger case for opting out than a brand using content primarily for awareness and brand-building, where being cited (even without a click) still carries real value.

4. Are you a news or developing-topic publisher? Given the new Top Stories carousel interaction, publishers in this category should weigh the loss of that specific placement carefully before opting out, since it may cost more than a standard AI Overview citation would.

What Most Experts Are Recommending Right Now

The general consensus forming among SEO practitioners is straightforward: most sites should stay opted in for now, at least until they have solid data showing AI Overviews are genuinely cannibalizing their traffic rather than supplementing it. The setting is also still rolling out gradually, UK-first, with global expansion planned but no confirmed date, so many site owners don’t yet have the option to test either way. Google has also indicated page-level controls (letting you opt out specific URLs rather than your whole domain) are planned for March 2027, which will eventually allow far more precise decision-making than the current domain-wide toggle permits.

Frequently Asked Questions

Will opting out of AI Overviews hurt my regular search rankings? No. Google has confirmed the setting only affects the three covered AI features and does not impact your ranking or indexing in standard search results.

Can I opt out of just some pages instead of my whole site? Not yet. The current version works at the domain level only. Page-level controls are planned for March 2027.

Does this setting affect the Gemini app? No. The opt-out covers AI Overviews, AI Mode, and AI Overviews in Discover specifically, not the standalone Gemini app.

Is this setting available globally yet? It rolled out UK-first under a CMA mandate and is expanding to a broader subset of site owners, but there’s no confirmed global availability date yet.

The Bigger Picture

This setting represents a genuine, if narrow, win for publisher choice in an AI-search era that’s mostly moved forward without asking permission. But exercising that choice wisely requires real data, not a reflexive decision in either direction. The smartest move for most sites right now is to monitor AI Overview appearances and their actual traffic impact closely, and make the opt-out decision only once the data clearly points one way, rather than guessing based on general anxiety about AI search.