Interchangeable Ad Formats: The Strategy Most Publishers Miss
By Loomit Team
Interchangeable Ad Formats:The Rabbit Hole Most Publishers Haven't Gone Down
Published by Loomit | Ad Monetization Insights
It Sounds Simple. It Isn't.
Interchangeable ad formats is one of the more interesting features in modern mobile adtech — and one of the most misunderstood. At its most basic level, the concept is straightforward: when a Rewarded Video (RV) placement opportunity arises, the system also considers bids for Interstitials (ITT). If an interstitial outperforms on CPM, it gets served instead, and the reward is still triggered for the user.
Clean and elegant in theory. But spend a few minutes thinking about the implications and you quickly realize how many decisions are buried beneath that simple description — decisions that most publishers have never had the tools to explore properly.
Two Objectives, Two Completely Different Strategies
The first thing to understand about interchangeable formats is that the objective you're optimizing for changes everything.
If you're optimizing for user experience, the calculus looks like this: even at the same CPM, you'd prefer to serve an interstitial over a rewarded video if the interstitial is shorter. You're trading ad length for revenue neutrality. If you take it further, you could quantify the revenue gains from swaps where ITT outperforms, and use that surplus as a 'budget' to also accept swaps at a slight revenue loss — netting to zero on revenue while consistently delivering shorter ad experiences. If done well, you print shorter ads for your users at no net cost.
If you're optimizing for ad revenue, the logic inverts: you only want to swap when it increases revenue. But this raises its own set of questions — and they get complicated fast.
What you're solving for matters. And most publishers haven't had the technology to explore the full decision space — until now.
The Layers of Complexity
Here's where interchangeable formats becomes genuinely intricate. Consider the questions that a rigorous implementation needs to answer:
On the user segment dimension:
Should you swap more aggressively for certain user segments? High-LTV users may benefit from shorter ads even if it costs revenue. Lower-value segments may have different sensitivities entirely.
Do users interact differently with RV opportunities when ad length decreases? Does reducing perceived length increase click-through or completion rates?
How does format swapping affect session times? If sessions are relatively fixed in length, does the time freed up by shorter ads get spent on more gameplay — and does that drive downstream engagement metrics?
On the network dimension:
Does knowing which network will serve the ad affect whether you should swap? If Network A consistently delivers 15-second interstitials and Network B delivers 30-second ones, should your swap logic treat them differently?
Is there a CPM threshold — a 'percentage to win' — below which you'd prefer to serve a rewarded video from Network X rather than an interstitial from Network Y, even if the ITT CPM is technically higher?
On the revenue optimization dimension:
If interstitials are 15 seconds each and rewarded videos run 30 seconds, does serving two interstitials in sequence effectively replace one rewarded video in terms of user experience cost? Would you accept that trade?
Could you exclude specific networks from being eligible for consecutive interstitial auctions to manage the quality of that experience?
Why Tech Is the Prerequisite
These aren't abstract philosophical questions. They're real optimization decisions that every publisher running both RV and interstitials is implicitly making — either consciously through strategy, or by default through inaction.
The reason most publishers haven't gone down this rabbit hole is simple: they didn't have the technology to explore it. Running this kind of multi-variable experiment — by segment, by network, by format, with clean A/B attribution — requires a platform built specifically for it.
You need the ability to quickly set up novel configurations, run them against specific user cohorts, and answer difficult questions at negligible cost. You need the speed to test multiple hypotheses in parallel, fail fast, and act on results before your competitive environment shifts again.
Increasingly, you also need someone — or something — to carry out these experiments for you: in real time, at scale, figuring out what works, when, and for whom. That's a lot for a person. Maybe an agent?
The Strategic Takeaway
Interchangeable ad formats is a case study in the gap between a feature and a strategy. The feature — allowing the system to choose between ITT and RV — is relatively easy to implement. The strategy — defining what 'better' looks like for each user segment, each network, each optimization objective — is where the real work happens.
The publishers who will extract the most value from this capability aren't the ones who enable the toggle and move on. They're the ones who treat it as an ongoing experimentation surface: continuously refining their swap logic, building institutional knowledge about which configurations work for which cohorts, and progressively automating those decisions as the data accumulates.
Loomit provides the technology to explore this space. The question for publishers is whether they're willing to go down the rabbit hole.
Want to see Loomit in action? Book a demo at loomit.ai
Related reading: Improve ad monetization with loomit