Traffic Discrepancy: Definition, Causes, and Ways to Handle them. Exclusive Analytics from HilltopAds

Stats don’t match up? We know firsthand how frustrating and confusing it might be. But don’t worry; the HilltopAds team will provide you with all the necessary instructions to sort it out once and for all. If you want to clarify what data discrepancies are, why they happen, and how to avoid and eliminate them, then stick to the very end!

Tracking discrepancy is the difference in data reported by two platforms monitoring the same ad campaign or traffic flow. Typically, these are tracking systems and ad network platforms.

Suppose that your ad campaign got AAA impressions in HilltopAds statistics. On the other hand, the dashboard inside the tracking tool says that the traffic volume is only BBB (BBB < AAA). Then, the tracking discrepancy will be calculated as the difference between them. For convenience, advertisers commonly use the Discrepancy Rate (DR, %):

When dealing with disparities, it’s important to remember that reaching DR=0% is impossible just because of the nature of this phenomenon. In our experience, DR=0% is only seen when the traffic source is full of bot traffic. However, there is a solid benchmark for the discrepancy rate that is considered to be acceptable: DR=30%.

Traffic discrepancy is a serious challenge for everyone involved in affiliate marketing. Since publishers only get paid for high-quality traffic, which is actually counted by our platform, lost traffic means missed monetization opportunities. Advertisers, meanwhile, must take into account the cost of lost traffic when planning their budget and calculating ROI.

Traffic disparities connected to RTB is a subject for another article due to its complex nature.

Finding the real cause (or number of them) of data mismatches requires time and expertise, but it must be done. Otherwise, you will keep losing money, clients, and time spent focusing on other aspects of advertising and monetization rather than trying to understand the data discrepancy phenomenon.

So, what may be the cause of it? There are different reasons for traffic discrepancies:

• A large amount of redirects.
• Incorrect postback setup.
• Tracking tool grouping non-unique impressions.
• Expired cookies.
• Fraudulent activities (e.g., bots, fake leads).
• Different traffic attribution models are used by the tracker and the ad network.

But not everything about this subject is straightforward. In fact, it’s more tricky than it seems. Let’s break it down. We identify two types of data discrepancies – “hidden” and “explicit”.

Handling significant discrepancies is one of the most critical tasks in optimizing both traffic buying and selling. That’s why we will go deeper into the approaches to resolve these issues.

Let’s master techniques that will help you detect data disparities faster before they result in unwanted money loss. At HilltopAds, we can identify them only after users report issues. We can’t prevent them earlier because external factors and behind-the-scenes tracker statistics remain hidden from us.

That’s why we encourage you to monitor the discrepancy rate closely from the very beginning of your campaign. Notice a suspicious anomaly? Contact your HilltopAds personal manager, and we will sort it out together!

Conseil d'expert : You can detect local discrepancies by analyzing specific traffic segments (e.g., GEOs). This approach allows you to spot problematic regions and then fine-tune traffic pricing specifically for this market, leaving the high-quality portion of your traffic at a high cost.

At this point, you probably already understand that resolving issues connected with traffic discrepancy is not easy. But don’t worry—HilltopAds has a few ideas for you on how to fight back!

To maintain accuracy in data reporting, please follow these recommendations from HilltopAds Chief Operating Officer!

Rimma

Chief Operating Officer

Pour les annonceurs :
• Set up tracking before launching the ad campaign.
• Run test conversion before starting an actual ad campaign.
• Monitor whether your domain (or any domain used in the redirect chain) is blocked by antivirus software or ad blockers. This approach can prevent you from losing visitors who can’t access your website due to active antivirus.
• Check the page loading speed. If it takes too long for a visitor to see a website’s content, the user will bounce, and the impression will not be counted in analytics. Lightweight pages are especially important in developing countries where internet speeds are often much slower.
• Minimize the number of redirects leading to your final offer. The more redirects you have, the less likely potential clients are to actually make it to the main offer page. This is another reason to work with direct traffic sources instead of resellers.
• Work with HilltopAds! On our end, the HilltopAds team constantly monitors domain reputation indicators from top antivirus platforms and Google Safe Browsing. It allows us to capture the moment when the ad might become problematic—not only for the publisher but for the advertiser as well. The more antivirus systems associate malicious activity with your offer, the lower the chance the audience will eventually see it.

Pour les éditeurs :
• Don’t try to mix in traffic from third-party providers. The most probable scenario is that you will get low-quality cheap traffic, which will fail to pass through filtering antivirus detectors.
• HilltopAds provides publishers with advanced anti-adblocker scripts that help bypass known filters and significantly reduce discrepancy rates.
• We purchase ad inventory across all GEOs and targetings from a publisher’s website. This ensures that the publisher receives a 100% fill rate with minimal discrepancy—something not always possible when working with third-party monétisation du trafic platforms.

Now that we have all the facts about data discrepancy let’s see how significantly this phenomenon affects affiliate statistics depending on the type of ad zone and traffic!

Nous avons décidé d'examiner dans quelle mesure les données relatives au trafic Popunder Desktop et Popunder Mobile peuvent varier en fonction de l'unicité des impressions. En d'autres termes, existe-t-il une corrélation entre la fréquence des annonces et le nombre d'impressions ? Examinons-le !

Pour notre étude, nous avons recueilli des données sur les zones directes "moyennes" et "grandes" sur une période d'un mois.

On the graph above you can see how the discrepancy varies depending on the share of unique impressions (i.e. impressions with uniquness = 1) of an ad in mobile popunder traffic.

• L'axe X représente l'unicité.

Note that a value of 10% is equivalent to the uniqueness of impressions in the range of 10% to 20% (compared to the whole volume of tested traffic at this stage), 20%—from 20% to 30%, etc.

• L'axe des ordonnées indique les valeurs d'écart.

À l'issue de cette partie de l'étude, toutes les zones peuvent être divisées en quatre groupes :

Par conséquent, les zones où la fréquence d'affichage des publicités est élevée ont tendance à présenter un écart de 1,5 à 3 fois plus élevé.

Le graphique ci-dessous montre la distribution des valeurs d'écart en fonction de la part d'impressions publicitaires uniques pour les zones de bureau Popunder.

Voici un point important de cette partie de la recherche. En général, contrairement au cas mobile Popunder, ces zones ne peuvent pas être divisées par l'unicité ou la divergence.

L'écart médian pour toutes les zones - à l'exception de celles dont l'unicité est extrêmement faible (comme le montre le diagramme de gauche) - reste compris entre 9% et 11%.

La seule dépendance notable est que plus l'unicité est faible, plus la limite des divergences est basse. Elle passe progressivement de 4% à 90%+unicité à environ 8% à environ 30% unicité.

Il existe un lien étroit entre la fréquence (unicité) des impressions sur une zone et les divergences dans les données des annonceurs. En réduisant cette valeur, vous obtiendrez de meilleurs résultats pour vos campagnes publicitaires, à moindre coût, ou vous augmenterez vos revenus de monétisation sans perte inutile de trafic.

L'importance d'un reporting précis des données est également soulignée par les résultats de notre recherche. Nous avons prouvé que dans les zones où les publicités sont affichées moins souvent (unicité d'impression plus élevée), les écarts diminuent de 1,5 à 3 fois. Cette dépendance est encore plus perceptible dans les zones mobiles Popunder.

Vous avez appris toutes les bonnes pratiques pour gérer les écarts de trafic. Quelle est la prochaine étape ? Commencez à les appliquer dès aujourd'hui - visitez votre compte HilltopAds et augmentez vos profits dès maintenant !

Si vous n'avez pas encore créé de compte, cliquez sur le bouton ci-dessous pour accéder au formulaire d'inscription !