Definition
Incrementality is the measure of how many conversions a marketing activity actually caused — conversions that would not have happened without it — as opposed to conversions it merely received credit for. It is established with a controlled experiment: a randomly chosen holdout group is not exposed to the activity, the exposed group is, and the difference in conversion rate between them is the incremental effect.
Everything attribution reports on top of that baseline is credit for demand that already existed.
Incrementality is the antidote to attribution's central flaw. A channel can show a large last-click or even multi-touch contribution while adding almost nothing incremental, because it is intercepting users who were already going to convert.
Brand search on your own name, retargeting of active users, and coupon sites are the classic examples. Only an experiment that withholds the activity from a comparable group can separate causation from coincidence.
In context
For iGaming, incrementality testing is how an operator resolves recurring budget arguments with evidence. Geo holdouts (pause a channel in matched regions and compare), audience holdouts (withhold retargeting from a random slice of eligible users), and PSA or ghost-ad tests are the usual designs.
Typical findings in this sector: brand-term paid search has low incrementality where organic already ranks first; retargeting of recently active players is largely non-incremental; coupon and "bonus code" affiliates convert users who were already at the deposit step; and upper-funnel content and prospecting, which attribution undercredits, often prove more incremental than their last-click numbers suggest.
The results feed directly into deal structure. If an affiliate's traffic is shown to be highly incremental — genuinely new players who would not have found the operator otherwise — that supports a higher CPA or better revenue share.
If a partner is mostly intercepting existing demand, the operator may cut the rate, restrict the method (no brand bidding, no coupon injection), or move to a model that only pays on demonstrably new customers. Incrementality is harder and slower to measure than reading an attribution dashboard, and the experiments cost some short-term volume, but for a large spender it is the only reliable way to know which channels are growing the business versus taking credit for it.
Worked example
An operator runs a four-week geo holdout on brand-term paid search. Regions with the ads paused lose only 4% of new registrations versus matched regions where it runs, because organic search captures most of that demand.
The operator cuts brand-search spend by 70% and redirects it to prospecting content, where a separate holdout shows a much higher incremental effect.
Related terms
Frequently asked questions
Browse the full iGaming & affiliate glossary — hundreds of EN/RU terms with examples.
← Back to glossary