#demand side platforms

Demand Side Platforms, Data Management Platforms and Exchanges

Client behaviour has evolved with time and much more focus has been put on audience reach and performance then just mere volume. Now when this change was observed by publishers, they realised that they could not keep 100% of their inventories with the sales team. So they kept 60% of their inventories with the sales team and 40% of their inventories were listed with something called as Exchangesor…

New Post has been published on http://www.learn-digital-marketing.com/facebook-overhauling-pmd-program-which-will-be-renamed-facebook-marketing-partners/

Facebook Overhauling PMD Program, Which Will Be Renamed Facebook Marketing Partners

Big changes are coming to Facebook’s Preferred Marketing Developer program, including an early 2015 rechristening as Facebook Marketing Partners, as well as a reorganization of partner companies by specific areas of expertise.

(more…)

I am happy to announce the Launch of Reduce Data Demand Side Platform [1] (DSP), an advertising platform that leverages big data technologies to deliver significantly optimized ad spends.

We believe that media buying is still inefficient and that there is a lot of room for optimization. Advertising fraud is more common than you think and this wastes billions of dollars of ad spending each year. 

Reduce Data is built from ground up using large scale data processing technologies (commonly known as Big Data technologies) to cut media waste, track campaign performance and optimize it at scale.  

Advertisers can reach users across desktop, tablets and mobile devices with Reduce Data.  We're rolling out the platform with a variety of new features and are extremely excited about the path ahead.  

Please do watch our blog or our Facebook page for announcements. 

--

This blog is an excerpt from our whitepaper: Optimizing Media Spends Request the full whitepaper. 

----------------------

Ad networks and DSPs have reporting solutions that can you understand how your money is spent.  Most solutions tend to provide enough insights for one to understand how your money was spent.  

However, there are several issues with these solutions because they may not help:

Check Targeting Criteria: Ad network analytics may not help you confirm that targeting criteria were met because the reports provided may be limited in their scope and detail.

Check Frequency Capping Criteria: Ad networks may not provide any mechanism to check frequency capping criteria and whether the traffic served had in fact met those criteria.

Click Fraud: Each ad network has its own definition of click fraud and that may not necessarily be right for you. Ad networks never provide insight into what clicks are actually fraudulent, how much of money was lost or recovered through click fraud and so on.  This is also the case with invalid or expired clicks which maybe from valid consumers but occurring long after the ad was displayed.

Viewable Impressions: It is estimated that 30% of all impressions are not seen by the users. This is a standard issue today in online channels and various third party technologies are coming up to address the issue.

In mobile, this issue is worse because aggregated supply of inventory gets sent to several networks and all ad networks respond with ad creatives without getting a response back as to which is actually shown. Some networks do measure the impression pixels back but not everyone does so.

In mobile however, this issue is getting resolved with the advent of Real Time Bidding through which an Ad Network or a Demand Side Platform (DSP) will exactly know which impression they have won and will be displayed.  But again whether the ad was seen or not is something that is debatable. 

In Mobile, Transparency & Brand Safety is a Big Issue:  The current market where inventory comes through aggregators has some serious issues with respect to the transparency of the kind of content/ sites / apps where the ads are served on. This raises some serious brand safety issues. While networks tend to lean towards the advertisers and try filter out content, systems are not completely mature enough to automatically capture adult / porn content and prevent an ad being served on mobile. Systems like AdExpose which provide such functions in online advertising, independently -  does not provide a comprehensive mobile service due to the fragmented nature of mobile devices and the limitations of technologies.

Even Among Online Ad Networks:  Not all networks are very transparent about where the traffic was received from and what kind of content was associated with your ads. 

The various scenarios of the traffic flow indicate a fast maturing marketplace with different partners trying to provide different value proposition.  However there are a few issues with the current arrangements:

Redundant Users and Repeat Ads: Advertisers typically add more than one network or DSP to the mix.  The problem with such a situation in mobile is that advertisers often tend to target and reach the same audience because there are a few large Aggregatorswho aggregate and supply the inventory the same users are often served ads from same advertisers despite frequency cap controls by various ad networks or demand side platforms (DSPs).  This leads to inefficiencies and wastage of spends.

Lack of Sufficient Data: While there are a number of players who offer publisher / app developer specific analytics, there are none who are focused on gathering data across the networks for analysis by the advertiser. Data such as device, country, carrier etc. are good for basic analysis and are served up by all networks.

But data relevant to campaign funnels, conversion attribution etc. are all missing in the mobile advertising space.

Lack of Real-Time Systems and Real-Time Data impacts Decisioning: Lack of real-time systems and real-time data impacts decisioning and predictive ad serving. Not all systems in the ad networks are real time and this as explained earlier has several undesirable effects. Frequency capping may fail because data sets are not updated at real-time. Real time data unavailability directly impacts the ability of the ad network to do proper real time decisioning and make the right bids at the right price. All of these directly affect ROI.

Ad networks use several predictive algorithms to decide which ad to bid or serve. These systems rely on data sets that are several hours old because it is not easy to feed all systems within an ad platform with real-time data.