Measurement Platforms Full Review


  • Type: Mobile Measurement Platform
  • Platforms: Mobile
  • Core Offering: Mobile attribution, audience builder, retargeting measurement, anti-attribution-fraud
adjust website

Adjust was founded in 2012 as Adeven, one of the first mobile attribution platforms. Adjust raised over $200M in their growth and recently was acquired by a major ad network – Applovin.

Adjust has always been very active in their opinions: •Using their own server farm (and not licensing a cloud solution) •Using an “Open Code” SDK •

Adjust had one of the most embarrassing server outages – not working for a few days, causing all reporting / postbacks to not be delivered to their customers and ad network partners servers crash

Adjust also had the biggest fraud problem out of all attribution solutions – but worst – they spun it off as if they were fixing a problem in the market, rather than admitting that they fixed a problem with their own SDK.

Unlike the other competing attribution solutions, adjust’s principle was an “Open Code” sdk.

Open Code – NOT Open Source.

adjust attribution

What happened ?

Adjust had no validation between their servers and the SDK – allowing anyone with basic knowledge of URL structure to trigger clicks and installs.

Adjust “fixed” it with an SDK version – but anyone smart knew that they could just insert sdk-version=x.xx and get around adjust’s “fix”.

Adjust have also made various comments about Apple’s IDFA deprecation saying that they spoke with Apple and that things will be different.

As time went by, it was clear that Adjust was unprepared to the new restrictions, as Apps featuring adjust SDK started getting kicked off the app store (which Adjust had to quickly fix their SDK and release a new one). Adjust claimed that the fingerprinting component still available in their SDK was an “old version”….

Adjust has repeatedly provided false statements to their customers and to the market.

Being acquired by an ad network just confirmed that Adjust is not an unbiased independent attribution provider and our recommendation was rather to avoid it at any costs! sdk spoofing fingerprinting sdk spoofing

Integration / Development Needed

Prepare yourself to do a LOT of development to get the adjust SDK integrated. Adjust has a ton of documentation on github , but the complexity of the open code SDK means that your product and engineering teams will need a lot of time to read through the documentation and work on the integration.

The adjust SDKs are full of issues and bugs. If you happen to stumble one of those you’re not lucky. 2 of our group members have been waiting for over 8 weeks with no resolution in sight. sdks


Once the extremely long and complicated integration process was done – onboarding was not as bad. Adjust does have a ton of documentation and a large customer success team that helps you understand the platform.

We don’t recommend going through the documentation – it’s just too much. As long as you are using a paid license, you should be able to get support in the onboarding process.

All of us (who actually used Adjust) liked the cohort reports. Those were very useful. cohort

Setting up events and new partners (if they are already integrated) was also relatively simple partners

The dashboard / report builder is decent, but we have seen better dashboards dashboard


Adjust has excellent support. They provide 24/7 support (online) , and you can also reach out to your customer success manager who can help you get around and navigate through technical issues.

The help center sends you to the online documentation area (which has a ton of documentation) – even too much. support support

Useful insights

If you want attribution reports – Adjust will have you covered. But if you want anything that resembles insights – prepare to pay more.

Adjust broke down their platform into multiple features, and most of the useful ones are “add ons” that you’ll need to pay for

Ridiculously enough – you’ll actually need to pay more for adjust to be protected against hacks (unbotify, fraud suite), pay more to track uninstalls and reinstalls, pay more to get skadnetwork handling, pay more for audience building, and more for cost reporting and marketing automation.

The only feature that is useful and comes with the costs of the attribution package is the cohort reporting. But even that does not provide insights – but only reports.

For almost $17,000 / month – we expected expected more add ons add ons


Adjust scored pretty low on accuracy (not the lowest, but not as high as you would like for your attribution provider). We noticed discrepancies of between 7% to 72% !!! discrepancies

Adjust has an article explaining the potential reasons for discrepancy. Here are only the titles from this article:

  • Downloads vs. Installs – Adjust tracks a “install” only once a user downloads, installs and opens the app for the first time. They explain that the discrepancies may come due to differences between “downloads” and ”installs” but both of these are not the case.
  • User based vs. device based installs – a user owning two iphones using the same account
  • Timezone and geo location
  • App update effect – it’s unclear how this impacts discrepancies
  • Third party stores installs
  • Comparing events / look back window
  • Fake user accounts
  • Reset of device ID
  • Differences between Facebook (as a self reporting platform)
  • Apple Search ads (self reporting platform)
  • Limit Ad Tracking (accounting for 20-30% discrepancy)
  • Google play vs. Adjust
  • Google adwords vs. Adjust

These discrepancies made adjust un-useable for us. Especially when discrepancies were so high.

We were really afraid to utilize the adjust marketing automation, as what this meant was that we’re letting Adjust manage our marketing decisions based on extremely inaccurate data. accuracy


SDK spoofing is adjust’s biggest problem. Anyone can easily fabricate fake clicks , installs and even in-app events on Adjust’s platform without putting too much effort.

You can actually buy “adjust installs for __________” (insert network name) for $3 per 1,000 installs in many hacker forums.

We found almost all of adjust’s largest customer names in these forums (i.e. “1,000 installs for miniclip campaigns on _________” )

Remember that Adjust doesn’t offer their “anti fraud” suite for free, but charges around $8,000 per month for this tool.

We did not like this. Our engineering team did not like this. Our finance teams did not like this. security

Privacy compliant

There are good parts and bad parts to the adjust privacy compliancy.

As the company was founded in Germany – Adjust used up most strict policies for data privacy. When GDPR was introduced, Adjust was the first one to comply with the law as per the requirements.

They did not store data for longer than a few days, and we had to digest the data from adjust and store it ourselves if we wanted to look back at user level data (adjust would only keep aggregated level data).

But when Apple announced the deprecation of IDFA and the ban over fingerprinting – Adjust seems to have not heard the message.

As of now (June 2021) Adjust continues to use fingerprinting in their SDK, simply naming it “probabilistic attribution”  and while it would be correct to say that all fingerprinting is probabilistic – not all probabilistic is fingerprinting.

Developers using the adjust SDKs have already experienced getting their app rejected.

We feel that using Adjust with iOS currently is a major risk to your business fingerprinting

Unbiased vs. Biased

Adjust ranks very low on this point. While it proclaims to be an independent attribution provider that is neutral and unbiased, the evidence plays very much against their messaging:

  • Adjust is owned by Applovin
  • Applovin is an ad network
  • Applovin is also the owner of multiple app studios

There is a high risk that your data is shared both with competing apps, as well as used to make the inventory of Applovin look significantly better than other ad networks.

The founders of adjust were also rumored to be investors in several DSPs and adnetworks.

  • Features like “bot detection” and anti-fraud are not part of the attribution pricing, but you would need to pay extra for those.
  • The feature to validate installs by getting app store “receipts” is also a paid feature

Adjust has over 3,000 ad networks they are integrated with, and even though, many of these ad networks provide almost nothing but fraud – Adjust continues supporting these networks and charging for their “anti-fraud” tools. biased attribution

Pricing Model

Remember how you could board a plane with a suitcase and choose your sit, but now you need to pay separately for every step in your travel experience ? That’s how adjust’s pricing model feels.

Adjust were smart by offering a “flat fee” package where you pay a flat fee and have no limit on the number of events or conversions tracked.

BUT – every feature you might want to add on top will cost you extra:

Do you want to build custom audience segments and test different creatives with split audiences ? There’s an extra charge for this.

Do you want to guard your ad spend budget against fraud? Filter out malicious activities that Adjust could easily block for you? There’s another extra charge for this.

You want to track uninstall events? Extra charge

Do you want to use the platform to manage your budget allocation? You guessed right – extra charge

Do you want to track install receipts from the app store, so that you have way less discrepancies? Prepare to pay more

Do you want to reduce the number of fishy bot traffic skewing your reports? This too – has an extra charge.

We calculated that a “full loaded adjust account” would cost as high as $50,000 per month.

Adjust’s pricing model is just like the printer’s model:

Pay a lot for the printer, and pay a lot more to continue being able to print by using the printer unique ink cartridges (that’s made from unicorn blood apparently).

The only similar company that uses this form of pricing is Ticketmaster pricing


In average, we ended up paying Adjust $21,548 per month.

Most of us used the flat package (about $16,500) and some of us “upgraded” to include features like Uninstall tracking, Fraud prevention, Control Center,  App store receipt.

Adjust was not the cheapest, but not the most expensive solution we reviewed.

For the level of discrepancies we found with Adjust – we feel that the price is unjustifiable – especially now that adjust was acquired for a billion dollars by an adnetwork – you would think that at least on Applovin inventory they would give some of the features for free – but no – Adjust charges a lot for the same services.

(note that two of us never got onboarded as the bugs we found during SDK integration were not “yet” resolved)

If adjust were in the business of selling cars… pricing

Review Summary:

Adjust was one of the first non udid mobile measurement platform (previously called AdEven). The approach taken by Adjust was using an open-code SDK, allowing developers to integrate the attribution SDK however they wish. Adjust customers suffered a lot from mobile-attribution-fraud as Adjust SDK was not protected , allowing fraudsters to simulate clicks and conversions very easily. Adjust has incredible documentation and their cohort reporting is simply the best. They were also the first attribution company to offer audience segmentation (used for retargeting) – a product that is slowly becoming redundant. Adjust has excellent support and documentation.

Adjust was acquired by AppLovin , one of the largest ad networks and mobile game publisher. Beware that using adjust means that you’re handing your data to an ad network and a gaming developer.

Pros & Cons:

  • Best documentation and an extremely simple on-boarding to the platform.
  • While the open code SDK is a nice idea, each of us needed engineering support to integrate the adjust SDK
  • Adjust is pretty solid in terms of data privacy – but it seems that their solution is using fingerprinting for attribution! (strictly forbidden and can place an app in jeopardy)
  • Fairly expensive, and price is based on the number of events tracked. In average, we each paid $21,548 per month (almost 0.3% of revenues), making the pricing model biased.
  • Even with all the marketing blabber – Adjust reported conversions that never happened. We found that those were fraudulent conversions.  None of us had a discrepancy of less than 5% (where the worst was +30%)

Measurement Platforms Review for

No Dev Needed★★★★★☆☆☆☆Privacy Compliant★★★★★★☆☆☆
Accuracy★★★★★☆☆☆☆Pricing Model★★★★★★☆☆☆
Useful Insights★★★★★★☆☆☆Price★★★★★★★☆☆