Reporting on cashback browser extensions and commerce media typically describes topline (gross revenue) results. But it doesn't help brands answer whether that revenue was incremental, or whether it would have shown up anyway.
Using a report on shopping browser extensions from CJ and Namogoo, plus Wildfire internal data, we’ve listed below several other ways to measure the value these channels create, beyond top-line revenue.
Brands should measure cashback browser extensions and commerce media using metrics beyond ROAS, including net revenue, repeat purchase rate, cumulative customer spend, new-to-brand rate, and incrementality. These metrics help reveal whether a campaign is actually creating incremental sales, increasing customer value, or simply receiving credit for purchases that would have happened anyway.
Discount and cashback costs show up as a clear line item, which makes it easy to just read them as eroding the margin. But, that line item is only half the picture.
In the CJ / Namogoo report, across 67 million shopping journeys, they found that cashback extension alerts only increased discount usage by 2.2%. But in parallel, these alerts increased the value of items added to cart by roughly the same amount.
Translation: discount spend went up, but average order value went up alongside it. Net impact on margin: close to zero. Net impact on revenue: significantly higher.
We found something similar in our own transaction data. We analyzed over 820,000 orders that passed through the Wildfire platform, and found:
the average order value on an extension-assisted purchase (e.g. where a cashback alert displayed to the user) was $144.
Meanwhile, the average commission a merchant paid was just $5.44.
This produced a blended CPA of roughly 4%.
Takeaway: the metric worth tracking is Net Revenue: AOV minus the discount or cashback cost. Compare AOV for sessions with an active extension interaction against the average discount percentage on those same transactions.
When the two datapoints move in parallel, higher discount volume isn't necessarily shrinking margin, it's actually contributing to higher-value purchases.
Extension users aren't a bargain-hunting segment that stops shopping when a company stops discounting. To the contrary, CJ's research describes them as "prolific shoppers" who actually spend 185% more on average than customers without an extension installed.
Wildfire's own data backs that up. In an analysis of over 770,000 individual extension users, almost half (44%) of extension-assisted purchases came from users who made multiple extension-assisted purchases in a year. They averaged 5.5 orders annually.
The combination of repeat purchase rate plus total spend across all purchases can act as a loose proxy for customer lifetime value. A full LTV model takes time and requires data most companies don't easily have access to.
But repeat-purchase rate and cumulative spend are simpler to calculate and answer a similar question: are these customers sticking around and buying frequently, or making a single deal-driven purchase and disappearing?
Often, it's the former.
Some advertisers running commerce media placements are trying to measure incrementality from their buys, and there are some alternate measures besides ROAS.
New-to-brand rate is one option: measuring whether a campaign converted shoppers who had never purchased from the brand before. With certain merchants, Wildfire can deliver a version of this signal.
In a sample of 3,000 orders Wildfire tracked, 44% of one merchant's and 39% of another's extension-assisted orders came from new or reactivated customers (with reactivations measured as they did not purchase in the prior 365 days).
Another option is incrementality testing conducted by the brand itself: using holdout groups, geo tests, and media mix modeling to isolate which sales actually happened as a result of a campaign, vs. trusting attribution data at face value.
These methods are much more rigorous, and can paint a fuller picture of incrementality. This is because a direct ROAS comparison will almost always under-value a channel that influences shoppers earlier in the journey, as standard attribution methods don't always fully credit upper-funnel touchpoints.
The common thread for measuring beyond simple ROAS
Each of these alternate measurement tactics informs something beyond pure ROAS. Net revenue can illustrate the margin impact of a discount, repeat rate and cumulative spend can be a proxy for lifetime value without creating a full LTV model, and new-to-brand rate or incrementality testing can show the lift a channel actually created rather than simply credit the revenue that happened to pass through it.
Extensions and commerce media both can influence a purchase decision before a shopper reaches a merchant's site, which is exactly why it’s possible that standard attribution methods or just looking at ROAS might undersell the full scope of their impact.
Measuring the additional value they create often requires looking past the transaction itself; these methods are a starting point for doing so.