Referral Links Clicked 3x More Than Banner Ads, Tracked
Affiliates in a mid-size U.S.-facing casino and sportsbook network clicked referral links at roughly three times the rate of banner ads over a 90-day window ending March 31, 2026, according to click-tracking data shared with iGaming trade press this week. The gap held across desktop and mobile web, though it narrowed sharply inside native apps, where banner impressions are often served in-feed rather than beside content. The numbers come from a single operator's affiliate stack, not an industry-wide census, which matters for how much weight anyone should put on them.
What the Tracking Actually Measured
The dataset covers 41 affiliates running a mix of comparison pages, picks newsletters, and short-form video placements for one licensed operator active in New Jersey, Pennsylvania, Michigan, and West Virginia. Over 90 days, the network logged 2.14 million tracked clicks across 6.8 million impressions, for a blended click-through rate of 31.5%. Referral links — the plain text-and-hyperlink placements, often inside a sentence or a table — accounted for 1.61 million of those clicks against 2.9 million impressions, a CTR of 55.5%. Banner ads, defined here as static or animated display units in standard IAB sizes, produced 530,000 clicks against 3.9 million impressions, a CTR of 13.6%.
That is a 4.1x raw ratio, not 3x. The 3x figure in the headline comes from a matched analysis: the operator's analytics team compared 22 affiliate placements where both a referral link and a banner ran on the same page during the same period, then averaged the per-placement ratio. That average landed at 3.08x, with a median of 2.6x. The spread matters. A handful of high-traffic pages with unusually strong text placements pulled the average up, and two placements actually showed banners outperforming links — both were sportsbook promos embedded in live odds widgets, where the banner sat inside content the user was already reading.
The distinction between 4.1x and 3.08x is not academic. If you're an affiliate deciding where to spend production time, the raw number suggests banners are nearly worthless. The matched number suggests they're a secondary channel that works in specific contexts. The second reading is closer to what the data supports.
Why the Matched Comparison Is the Honest One
Cross-placement averages are contaminated by page quality, traffic source, and audience intent. An affiliate with a well-ranked "best Michigan casino bonuses" page will post high CTRs on everything it runs, including banners. Comparing that page's links to a banner on a low-traffic blog post tells you about the pages, not the formats. The matched set controls for page, audience, and time window, which is why the operator's own team led with 3.08x rather than the flashier raw figure.
Text Links Win on Intent, Not on Design
The obvious explanation is that referral links sit inside sentences people are already reading, while banners sit beside them. That's true but incomplete. The click data shows a pronounced pattern by placement context:
- In-body links inside editorial prose: 61.2% CTR across 890,000 impressions. These are the "our top pick for fast payouts is [brand]" placements.
- Comparison table links: 48.7% CTR across 1.4 million impressions. Lower than prose, higher than anything else.
- Sidebar text links: 22.4% CTR across 610,000 impressions. Roughly half the rate of in-body links, suggesting position within the reading flow matters more than the format itself.
- Standard banner units (300x250, 728x90, 160x600): 13.6% CTR blended, with the 300x250 performing best at 17.1% and the 728x90 worst at 9.3%.
The sidebar number is the interesting one. A sidebar text link is still a referral link by format, but it performs closer to a banner than to an in-body link. That undercuts the tidy "links beat banners" story. What actually seems to drive clicks is adjacency to content the user chose to read, and referral links are simply easier to place there without wrecking the page.
There's a second factor the data can't fully separate: disclosure. Most of these affiliates label referral links with some variant of "partner" or "advertising disclosure," often inline. Banners frequently carry no label at all, or carry one in a footer. It's plausible that clear inline labeling increases trust and therefore clicks, which would be an uncomfortable finding for anyone arguing that disclosure kills conversion. The dataset doesn't isolate this, and the operator was candid that they haven't tested it.
The Mobile Web Split
Mobile web produced the widest gap: 3.7x in the matched set. Desktop came in at 2.4x, and in-app traffic at 1.6x. In-app banners are frequently served as native units inside a feed, which makes them functionally closer to a text link than to a traditional display ad. Once you strip out native in-feed units and look only at standard display sizes inside apps, the matched ratio climbs back above 3x. The format comparison is really a placement comparison wearing a costume.
What This Doesn't Tell You
Click-through rate is the cheapest metric in affiliate marketing and the least connected to revenue. A referral link that gets clicked because it's ambiguous — "check current offers" with no brand named — can post a huge CTR and convert terribly, because the user arrives at the operator's page expecting something else. The operator in this dataset declined to share conversion or revenue-per-click figures, citing partner agreements. Without those, the 3x claim describes attention, not value.
There are three other limits worth naming:
Sample bias. Forty-one affiliates, one operator, four states. The affiliates who agreed to share tracking data are not a random sample; they're the ones confident enough in their numbers to hand them over. Affiliates with weak link performance had no reason to participate.
Attribution overlap. A user who clicks a referral link on Monday and a banner on Wednesday may be counted twice. The operator says it deduplicates by user ID within a 30-day window for conversion purposes but counts raw clicks for CTR. That's standard, and it inflates both formats equally, but it means the absolute CTRs are not clean measures of unique user behavior.
Survivorship in placement. Affiliates tend to put referral links on their best pages and banners on the rest. The matched analysis addresses this, but only for the 22 placements where both formats coexisted. Those 22 pages may not represent the other 19.
If the industry wants a defensible benchmark, someone needs to run a controlled test: same page, same traffic, randomized assignment of link versus banner, tracked through to deposit. Nobody in this dataset did that. It's the kind of test an operator could run in two weeks and almost never does, because affiliate relationships are negotiated individually and nobody wants to tell a partner their banner is dead weight.
Where the Money Actually Moves
Revenue-share deals dominate U.S. affiliate contracts in the regulated states, typically 20% to 35% of net gaming revenue for the first 12 to 24 months of a referred player's life, with some deals running longer. Cost-per-acquisition deals run $150 to $400 per depositing player depending on state and product, with sportsbook CPA at the low end and casino CPA at the high end. Under rev-share, click volume is a leading indicator at best. Under CPA, it's closer to irrelevant — the operator pays on the deposit, not the click, and a high-CTR placement that attracts bonus hunters costs the affiliate money in wasted traffic.
That asymmetry explains why the 3x finding will get quoted in sales decks and ignored in contract negotiations. Affiliates don't get paid per click. They get paid per player who deposits and, under rev-share, keeps playing. If referral links attract a different kind of user than banners — more deliberate, less impulse-driven — the revenue per click could be higher or lower than the CTR ratio suggests, and the dataset here can't say which.
There is one place the CTR gap has direct financial consequences: paid traffic. Affiliates buying search or social traffic pay per click regardless of format, and a placement that converts clicks at 3x the rate changes the math on what they can afford to bid. A handful of the 41 affiliates in this dataset run paid acquisition, and their internal numbers, if they exist, would be more useful than anything in this report.
The Question Nobody's Asking
The finding that referral links outperform banners is not new. Affiliate marketers have said some version of it since the first text ad outperformed the first banner in 1994, and the underlying reason hasn't changed: people click things that look like content and ignore things that look like ads. What's new here is a number attached to it from a licensed, regulated U.S. operator rather than an offshore program with no reporting standards.
The more useful question is whether the gap survives when you measure revenue instead of clicks, and whether it holds for the affiliate tiers that actually drive most of the volume. The 41 affiliates here skew mid-size. The top five affiliates in most U.S. programs operate at a scale where they negotiate custom placements, run their own creative, and don't use standard banner units at all. Their economics may look nothing like this sample.
Until someone publishes deposit-level data by placement format, the 3x figure is a reasonable argument for writing better in-body links and a poor argument for deleting your banner inventory. The operator that shared this data is already running the test that would settle it. They just haven't said whether they'll publish the results.