In our recent programmatic webinar, “Tricks of the trade: holiday edition” we received a lot of great questions about how to improve campaign performance. In this blog post, we’re answering your questions.
Many high-impact units have significantly reduced load times and heavy discrepancies in viewability reporting. Can you explain why this may happen?
This could be due to several factors. One of the more common factors is that the ad server or SSP that is being leveraged cannot accept creative that has such a large file size. This could cause a discrepancy in viewability reporting if viewability on the publisher or SSP side is being measured at ad unit load vs. creative unit load. In laymen’s terms the creative is likely never loading. This is fairly common throughout the industry, especially with video creative which can be served in HD to player that is not setup to accept HD ads.
Can you review what SIVT is? And how it’s measured?
SIVT stands for Sophisticated Invalid Traffic and it is the most prevalent form of ad fraud. In order to effectively detect SIVT, advanced analytics, multipoint corroboration, and significant human intervention is required.
Examples of SIVT include:
- Falsely represented sites or impressions
- Hijacked devices: a user’s device (browser, phone, app) is modified to request HTML or make ad requests that are not under the control of a user and made without the user’s consent (for example, operations made by a bot)
- Hijacked sessions within hijacked devices
- Hidden/stacked/covered or otherwise intentionally obfuscated ad serving
- Anonymized proxy traffic
- Incentivized manipulation of measurements such as payment for video interaction or guided browsing
- Misappropriated content
- Falsified viewability measurement
- Cookie stuffing
- Manipulation or falsification of location data or related attributes
How is SDK distribution scaling? How much improvement are you seeing in the availability of viewability-measureable in-app inventory? Where can the best scale be found?
While we can’t speak to all SDKs, IAS’s ability to scale has been greatly improved by partnerships with MoPub and InMobi, the two largest mobile SSPs in the marketplace. Their publishers have rapidly adopted this solution, significantly growing scale. Today you can access measurable inventory through either of these SSPs via a deal id or a flag in the bid stream, although the ability to ingest a flag varies by DSP.
Why is the best inventory available on “return week”?
Consumers are back at work, and browsing more pages on the internet versus the holidays when they are spending time with friends and family. Additionally, ad spends decrease significantly which takes away the incentive for bad actors who are running bot fraud schemes.
Why do “use it or lose it” budgets increase risk of fraud on holiday seasons?
Many times programmatic buyers will remove all pre-bid targets or filters to increase their scale at the tail end of the campaign because they know that they are going to lose their budget anyway. This allows fraudsters or bad actors to capitalize on an unprotected spend.
Could higher fraud and lower viewability on key holiday days be due to traders being forced to source from lower quality inventory sources in order to achieve scale?
It could be. It could also be that the fraud levels are fairly consistent with the exception of a few holidays like New Year’s Eve, but due to low impression volume, fraud makes up a higher percentage of the inventory. The moral of the story is that you just need to make sure you’re protecting your investment by applying pre-bid filters, or at a minimum, reporting on your programmatic ad buy.
Should days of poor delivery be avoided altogether even if when pre-bid targeting is enabled?
No, you don’t need to avoid these days. You just may need to adjust your bidding behavior to win viewable and brand safe impressions.