At Thesis we perform new ad account audits nearly every day. To aid those conversations, I've maintained my own internal ROAS benchmarks for reference. In this post, I've sanitized that ROAS data so that I can share it for general consumption!
Adam Lovallo
February 17, 2021
At Thesis we perform new ad account audits nearly every day. To aid those conversations, I've maintained my own internal ROAS benchmarks for reference. In this post, I've sanitized that ROAS data so that I can share it for general consumption!
What this data is:
Prospecting ROAS data for 30 direct to consumer/ecommerce brands spending in the US & CA.
I've deliberately only pulled prospecting campaigns (those ad sets that exclude site visitors and purchasers). I find that different brands' spends allocations for retargeting vary dramatically (see our blog post on that subject here), that retargeting often has massively different performance when measured on an incremental basis (see our blog post on that subject here), and that retargeting campaign structures (ex: retargeting existing purchases or not) vary widely from brand to brand. For all of those reasons, I find a prospecting focused comparison is most useful, though I think other approaches to benchmarking ROAS could also be very useful (especially on a blended ROAS or "market efficiency ratio basis... perhaps a topic to address in a future post)!
All of the brands included are focused on driving conventional, non-recurring purchases (although some of these brands have subscribe and save options). I've excluded any brands that are primarily subscription focused.
Purchase & revenue data is pulled from Facebook using a 7-day click attribution window (their new default attribution window!)
I pulled this data from Feb 1, 2021 to Feb 9, 2021 so that the data in question is fully attributed (aka the full 7-day attribution window has elapsed for each day)
I've rounded spend data up to the nearest ten thousandth
I've rounded the AOVs to the nearest tenth and I've further categorized the higher AOV products into ranges ($100-$250, $250-$1000, >$1000).
Lastly, I want to caveat that: A) these are only 30 data points B) obviously prospecting ROAS is not the goal... blended ROAS is the real goal! C) different categories have different repeat purchase rates and thus may have different goals D) different categories might need to be looked at with different attribution windows E) I'm sure there are a lot of additional holes to poke in this data!
Benchmarks
There it is! The average prospecting ROAS on a 7-day click basis is 1.10 and the median is 1.04. Based on everything I've seen in the last 6 months that honestly feels about right for D2C. I'd love to hear if anyone out there disagrees or can suggest a better way to benchmark this!