Incrementality Platform Reviews
Remerge is a mobile retargeting dsp based out of Berlin. Founded in 2015 by 5 co-founders coming from Madvertise (an early mobile ad network)
Remerge raised $4M from investors, including Point Nine capital, IBB, and various angel investors from the adtech industry.
Remerge currently has over 150 employees, mostly in their Berlin office.
The reason I wanted to review remerge, even though they are a retargeting dsp (i.e. media vendor) is that remerge has been offering “cost per incremental conversion” as a unique offering of their media sales services.
Cost for Incremental Conversion sounds like a dream coming true!
If Advertisers can only pay for conversions that were driven by their advertising, and not pay for conversions that would have been generated without Advertising – that would make every advertiser grow in absolute terms and in profitability!
But dreams coming true doesn’t mean they are good dreams.
I’ve tried remerge a few times in the past. Mostly because friends in the industry told me that the service is good and remerge generates high quality of results for retargeting.
What my friends failed to notice/realize/know is that it’s very easy to generate good results in retargeting – if you are already targeting users that would have likely to convert themselves without advertising!!!!!
I ran a few campaigns with remerge, even though I knew that their “cost per incremental” conversion was completely off – just to understand their approach.
It’s very simple – Remerge uses “hold out groups” (i.e. a control group and a test group).
Remerge serves the test group with ghost ads / ghost bids, and the control group with my real campaign ads.
The outcome of the campaign is measured, divided into the groups, ending up with a graph that looks something like this:
With this approach, incrementality is very easy to calculate:
Test (conversion rate) – Control (conversion rate)
Test (conversion rate)
What remerge fails to show and explain is that if their solution truly provided incremental results – the results for unexposed group should have been ZERO
Like a bad magician using a good looking assistant – Remerge invests a lot into marketing to come up very scientific – but all of this is just to deflect advertisers from the obvious.
Remerge makes its money from advertisers by selling them media. Naturally, their incentive would be to show positive results from their “incrementality tests”
what more is that their incrementality tests and insights only represents the campaigns running through their own platform. Now if you as an advertiser only work with remerge their results have some chance of being thorough, but given that remerge is just one of a hundred retargeting dsps – no advertiser is likely working exclusively with remerge to the point where their results are valid.
There’s nothing wrong with doing incrementality testing using an audience split. If you are in 2018. But as our world moved on towards privacy and creating “audience groups” is not possible anymore.
My only assumption is that remerge keeps selling incrementality campaigns to naïve customers who don’t know about the changes in regulations, only to target whoever they want and charge unsuspecting clients for what they report as “incremental results”
I can even prove this by pointing out that remerge only talks about “incremental ROAS” which is NOT the same as “incremental results” or “incremental ROI”
Incremental ROAS comes to calculate is the return on spend by calculating the revenues generated by users attributed to the campaign.
But what if the users attributed the campaign would have been attributed to organic results if the campaign did not exist ?
If that is the case – even a campaign “positive ROAS” can actually influence ROI negatively
Don’t fall into someone’s explanations if their agenda is to get you to spend money with them!
How easy is it to integrate with Remerge ?
It’s really easy to integrate data from MMPs to Remerge. They have an integration module for all attribution solutions, including adjust, appsflyer, kochava, branch, singular, and mparticle.
What is unclear is what is the approach for incrementality tracking with audiences when audiences are no longer possible.
Ease of integration: ★★★★☆
What is Remerge pricing ?
Remerge sells media on a CPM, CPC, CPI, CPA, and also offering “incremental cost per conversion” (iCP-conversion?)
As their interests are to get customers to spend more with them, the pricing is extremely unbiased – hence, I could not recommend it.
What about the insights ?
The insights are not so useful, as it’s like the clerk in the store telling you that “you look great in those jeans” even if you look like a hobo…
Remerge sells media, and their interest is to get me to spend more.
I did not tell them that I’m testing them just for the review.
“we hope you are happy about the results!” said Scott H, my “customer success manager”…
Useful insights: ★☆☆☆☆
Were the results from Remerge unbiased ?
Unbiased results: ☆☆☆☆☆
Privacy Restrictions compliancy
No big surprise here. Remerge scored at 0.
The whole approach of incrementality testing for remerge depends on creating a split audience test.
This is the opposite of compliance with the latest changes in privacy regulations….and most importantly – is NOT possible.
I have a suspicion that they might be doing some funny business….
Compliancy with privacy requirements: ☆☆☆☆☆
What is Remerge overall score ?
Remerge sells media. I wanted to review them, as they have the most unique offering: cost per incremental conversion or cost for incrementality. That’s a huge promise to make.
Unfortunately Remerge does not cash in that cheque.
The “incremental” results are only showing what Remerge are comfortable with showing.
There’s no real measurement and there is no real incrementality.
I was able to easily check that the “incremental conversions” were mostly cannibalizing my own organic conversions.
I ended up running a couple of campaigns for the sake of the review. Learned how “great the results were” from the customer success manager I was assigned with, and that’s that.
I’ll stick to media vendors where can control all aspect of trading, and leave incrementality testing to platforms that are actually unbiased like MetricWorks and INCRMNTAL.
Incrementality Platform Reviews rating for Remerge:
- Ease of integration: ★★★★☆
- Pricing: ★★☆☆☆
- Useful insights: ★☆☆☆☆
- Unbiased results: ☆☆☆☆☆
- Compliancy with privacy requirements: ☆☆☆☆☆
- Overall: ★★☆☆☆
About the author:
Michael Holmstein, experienced marketer with over 7 years experience in app marketing. I’ve worked at gaming companies like MZ and Playstudios, and fintech companies like Current and Coinbase. Attribution and incrementality measurement is a personal passion of mine.
By: Scott H.
Hey Michael, It’s Scott. I’m really sorry that you felt that the results were below your expectations.
I would be happy to schedule a call with you and my team lead. We are very eager to hear what we could have done better.
If you could also remove this review in the meanwhile, that would be much appreciated.
By: Michael H.
Hey Scott, sure, I’ve reached out for a call. But the review remains on. If you disagree with anything written here, you are free to voice it in the comment section.
By: Philip M.