Data drives business decisions. This is why ambiguity in it can cause problems. A difference in analytical data between two publishers and advertisers, ad discrepancies occur due to data mismatch. However, that is just the surface. The implementation of different formulas, tools, and techniques to calculate analytics data causes these advertising discrepancies.
What Is Ad Discrepancy?
To put it simply, it is the term used for data inconsistencies that occur while measuring data. The number of ad impressions between the publisher’s ad server and the demand partner’s ad exchange or ad server. It could also mean the difference in ad requests between the ad servers of both publisher and partner. This problem due to data mismatch is quite complex. Impression measurement depends on factors like site loading time, ad-blocks, and user interactions. When this number has a mismatch, you get an ad discrepancy.
The reasons for this happening are many, the most common being the mismatch in impressions because the sell-side and buy-side use different counting methods for impression measurement. Other common reasons for this include losses due to exchanges between the publisher’s and partner’s ad servers, redirects between the landing page and the ads, network connection failures, and server timeouts amongst others.
Right now, the two kinds of discrepancies being faced by the ad industry are –
- Between ad tech vendors.
- Between sell-side and buy-side platforms.
These two may mix together and create further issues because publishers and advertisers have different standards for ad impression and delivery data calculation. Publishers might depend on the number of ad requests generated to calculate impressions, while advertisers may depend on the number of ads delivered.
How Do You Calculate Ad Discrepancy
Usually expressed as a percentage, ad discrepancy can be calculated as follows –
(Difference between the number of ad impressions recorded by a publisher + number of ad impressions recorded by the ad exchange)/average of two numbers further multiplied by 100. This way you can see both you and your partner’s discrepancy, i.e by how much % your and your partner’s number differs from the actual value.
How Does Ad Discrepancy Occur
There are quite a few causes of ad discrepancies. Some of the main ones have been listed below.
The total time a page takes to load is critical in the programmatic auction or real-time bidding. It is impossible to have a page load instantly. Distance between a hosting server and user and ad server latency are a couple of factors that affect page loading time. High page latency often leads to unfulfilled impressions. Users leave pages if ads load after the page content. The request may be recorded, but there’ll be no impressions. Moreover, if users abandoning the page, your bounce rates will increase, damaging your SEO.
With around 27% of users using ad blockers in 2021, fewer and fewer ads are making it to them eventually. This leads to ad request discrepancies between the servers for you and your partner. You get neither ad delivery nor measurement, as ad blockers block both. Sometimes the odd domain or ad could go through, creating an anomaly. Report figures and tracking codes are severely affected by ad blockers which show different values on you and your partner’s analytics report. For example, if an ad blocker stops all Ad Class, then the user’s browser will not show any ad placed in said Ad class. However, the publisher still sees the user as active. However, they can’t display any ads to the user and get any impressions.
There is a fair share of human errors that cause ad discrepancies. If publishers don’t correctly integrate ad codes or place an ad code wrongly, things can go awry. For example, placing a video ad script outside the video player instead of a banner ensures users don’t see ads, nor do you get any impressions, which should be avoided. Always cross-check all your actions when working with ad codes. Ad networks give special instructions for publishers, like step-by-step recommendations on ad code installations. If you want accurate ad impression and stats tracking, integrate your ad code correctly.
Another overlooked reason for ad discrepancy, some creatives are just heavier than others. It is known that video ads are heavier than display ads, requiring more time to load. In the case of out-stream video ads, for instance, if a user visits your page but leaves before the ad loads, you’ll see an anomaly in the final number of video play impressions. The publisher’s ad server would’ve counted the impressions, without the ad being displayed at all.
Time Zone Differences
Most reporting tools don’t provide real-time data. Take, for instance, an advertiser using an EAT time zone. If he is working with a site owner who is in the EC time zone, the latter will be generating reports by 10 pm daily. This causes a 2-hour gap. Because time zones will never change, this inconsistency in data reports on both sides will always be there.
What Do You Ultimately Do
You can’t totally get rid of ad discrepancies. However, you can aim to keep that percentage below 10. Here are some tips.
A browser cache saves details of website visitors, so when they come to the site again, the page loads faster and smoother. Ad units get attached, and if the ads are not updated regularly or loaded accurately on the browser, this affects analytics reports. Cache busters are tools that ensure that each time a person visits your site, the ad server loads ad units, instead of the web browser cache doing so.
Traffic Validation Tools
Certain traffic validation tools and platforms can help to owe to their bot filtering feature, scanning, and removing invalid traffic.
Try To Coincide Time Zones
This might not be so easy. Publishers and advertisers need to decide on a time zone in which both parties can perform analytics.
Use Common 3rd Party Platforms
While calculating audience behavioral metrics, both publishers and advertisers could use the same 3rd party platforms across both accounts.
Daily reports containing collected user data, updated by publishers and advertisers can help you detect the smallest change and keep a close eye out for discrepancies.
Whitelist Your Site
Your ad partner needs to whitelist your site. A white list is a list of sites that an advertiser can work with. Any site not on that list won’t be able to display ads.
In conclusion, you have to keep an eye out for anomalies in ad statistics. Whether they come from the ads, publishers, advertisers, or the ad network, find that discrepancy and see whether it has happened before. Finding the problem is the first step, and if you can’t pinpoint it, work closely with your team in combing data.
Always keep running tests and create a detailed impression discrepancy report, which can help you narrow down the causes for ad discrepancy. Publishers always need to be aware of where impressions are lost and try to find ways to work around the problem. Read our blog on striking the right balance between ads and user experience to give your site visitors the browsing experience they’ll remember.