The MMP buying guide – Your handbook for getting the foundation of your tech stack right

Introduction

If you’re here, you probably already know you NEED a mobile measurement partner, also known as an MMP. Data-driven growth strategy is your holy grail and mobile measurement is your map to the stars. Obviously.

Being able to answer mission-critical questions such as “which user acquisition (UA) channels bring in the highest lifetime value (LTV) users?”, or “which campaigns are yielding the highest return on ad spend (ROAS) and in which regions?” — are but a fraction of the real-time insights an MMP solution can offer.

But where to start?

You already know that with knowledge comes power. But how DO you choose an MMP?

The current MMP market is made up of a relatively small number of players, attesting to the heavy tech and thousands of integrations that are required to allow it to run.

Unlike other tools in your tech stack, the MMP market is pretty clearly defined. Each player comes with their own set of strengths and compatibility with different business sizes and needs.

And while at first glance, it may seem like all the solutions out there are quite similar — i.e. clicks measurement, click-to-install attribution, in-app events attribution, etc. — there are big differences when it comes to the needs, size, and maturity of your company.

Quite a few important factors need to be taken into consideration when on the prowl for the right measurement partner — especially in the face of the recent and future privacy changes, courtesy of Apple and ֿֿֿֿGoogle.

So how do you determine which MMP is the best partner for your app’s unique needs?

You needed answers — we listened, and put together all the essentials: how to identify and pinpoint an MMP solution that meets your needs like a snuggly glove, which pitfalls to avoid, and how to get crucial buy-in from your boss.

Chapter 1

Times they are a-changin – The MMP landscape and why privacy in our post-iOS14 era is such a major consideration

The major MMPs on the market are AppsFlyer, Branch, Singular, Adjust, and Kochava. All aim to support slightly different needs and business sizes.

What turned this entire industry on its head and redefined mobile attribution as we once knew it — was iOS 14.

Prior to Apple’s shocking announcement in April 2021, MMPs were able to access iOS users’ ID called IDFA and use them in the attribution process to deterministically attribute with great accuracy.

Post iOS14, however, IDFA access was made available only given user consent, with Apple’s SKAdNetwork framework controlling the attribution process on an aggregate level.

What does that mean for mobile attribution?

It means that both MMPs and ad networks get their iOS data from one source — SKAdNetwork (SKAN), which comes with its long list of limitations.

These limitations, however, can be mitigated with the right MMP by your side. Here’s the gist of it:

Challenge: Data flow is decentralized and distributed, which means each network receives its own postback data.

Solution: A consolidated, unified single source of truth (SSoT) that can accurately dedupe multiple sources of data .

Challenge: Attribution flows in SKAN are limited to direct app-to-app and click-to-install.

Solution: Aggregated advanced privacy (AAP) and web-to-app measurement.

Challenge: There’s an unavoidable fraud risk as data could be easily manipulated en route.

Solution: Fraud detection and prevention.

Challenge: Even with the release of SKAN 4.0, KPIs are still limited and timing constraints still exist, meaning the full measurement lifecycle is limited to three postbacks, each based on a specific activity window (0-2 days, 3-7 days and 8-35 days).

Solution: Predictive analytics.

Challenge: SKAN’s conversion values are extremely complex to figure out and configure.

Solution: Conversion studio.

Challenge: When Apple’s privacy threshold isn’t met, conversion values come back as “null”.

Solution: Machine learning-based model that provides accurate results down to the adset level (more details on this ability will be announced soon).

So in case that wasn’t clear, privacy has taken center stage in app marketing, impacting attribution models, campaign optimization, monetization methods, remarketing, and mobile fraud.

Apple, by the way, is most certainly not the only contributor to this massive shift towards user privacy. Meta also no longer provides advertisers user level data on Android (only to MMPs), Google is deprecating 3rd party web cookies, terminating GAID (its user identifier) and is rolling out the much-discussed Privacy Sandbox.

These, together with additional changes that are yet to come, are all adding an undeniable layer of complexity to the ability to run an end-to-end, granular mobile measurement.

The good news for you is — there are MMPs out there that stay ahead of the privacy curve to ensure you can still gain visibility into mission-critical trends, draw timely and actionable performance insights, carry out a data-fueled growth strategy, and optimize your budget spend on the fly.

Which brings us to the more practical part of our guide.

Chapter 2

Use cases – Who uses an MMP solution and how?

An MMP is a marketer’s best friend, that much is true. But, the wealth of real-time, accurate, cross-channels data it offers can and should be used by stakeholders outside of marketing as well.

Let’s explore a few use cases for what an MMP offers each team:

Use case #1 – Marketing team

Measure and optimize ongoing campaigns.
Use a single source of truth for your siloed performance measurement (this includes accurately deduping data from multiple sources, data inconsistencies, unstandardized & messy data coming from partners and ad networks, unattributed non-organic installs, evaluating actual iOS cost metrics, and understanding organic baseline).
Calculate user LTV, spend, and ROAS data across all partners, platforms, and channels.
Create smart audiences to be able to craft effective and highly personalized campaign segmentation.
Allocate budget effectively according to real-time, accurate, and fraud-free campaign performance.
Block fraud from polluting your data and draining your budget.

User Acquisition team

Create a delightful customer experience using deep linking to streamline journeys like web-to-app, email-to-app, QR-to-app, social-to-app, referral-to-app, text-to-app, CTV-to-app, etc.
Gain access to aggregated user-level data using data clean rooms, and then draw actionable insights around UA activities.

Use case #2 – Product team

Test, measure, and optimize new feature adoption / product updates, and keep track of where users are getting stuck by monitoring user journeys.
Gain important insights around UX by putting in place in-app feedback surveys that will be shared with users during pivotal stages (e.g. Day 1, Day 30, etc.), and later on funneled into periodical product updates.
Understand how different cohorts of users interact with your app, and adjust product strategy and roadmaps accordingly.

Use case #3 – BI/Analytics team

Stream data into internal BI systems with ease and on the spot.
Use predictive modeling to examine historical campaign data, past user behavior data, and transactional data — to predict future actions.
Enable you to provide the marketer with the necessary data — so the latter can nip underperfoming campaigns in the bud, or quickly double down on successful investments.

Use case #4 – R&D team

Simplify integration and compliance for complex tools such as Apple’s SKAdNetwork and Google’s Privacy Sandbox.
Save endless hours of connecting to each media partner’s SDK individually.
Integrate with other tools in your tech stack with ease.

Use case #5 – Security team

Identify, mitigate, and prevent ad fraud using real-time fraud detection (consisting primarily of bots, click flooding, device farms, and click injections).
Ensure full compliance and alignment with all applicable privacy legislation and platform requirements, including authentication security measures.
Enable business continuity by relying on 99.9% server uptime and uninterrupted data availability.

Use case #6 – Customer retention team

Keep churn rates at bay and improve user LTV by continuously monitoring retention and churn rates.
Monitor granular data around when and why users are leaving the app, which will also inform your marketer’s decision-making around in-app events.

Chapter 3

How to sell an MMP solution to your boss

Let’s say your boss is not entirely sold on why you need to partner with an MMP to begin with – How do you respond?

First thing’s first. You need to think about all your current and potential blind spots, how they affect your ability to measure accurately, and in which ways do these blind spots affect your business as a whole.

How do you pinpoint these blind spots? Start with the following questions:

Am I able to accurately assess how my campaigns are performing? How is their performance affecting other mission-critical activities in my organization?
Can I look at my performance measurement across channels, devices, and platforms as a whole with full trust?
Given the increasing user privacy wave, do I have the right technology layer in place to help me measure granularly while safeguarding my customers’ data?
Can I really drive growth with confidence while flying blindfolded?

Assuming the answer is “no” to most if not all of the above, the need for an MMP solution should now be clearer to your boss. After all, each and every “no” ends up costing your business money.

But if your boss is still not 100% sold, here are two key points they won’t be able to argue with:

Point #1 | Your business has to have the right data

The more data the merrier, right? Wrong.

In today’s day and age, it’s really not about an abundance of data anymore. It’s about the right data.

When you think about technology, any technology, the intended purpose behind it is to solve a business problem.

So when you talk to your boss about the MMP space and why you need one, bear in mind that this is all about creating an enablement layer of mission-critical information available at your fingertips, that can then influence and solve your business challenges.

What do we mean by the right data? The right MMP offers:

Cross-channel user journey measurement – So you can connect the dots in a fragmented reality.
Accurate LTV, ROAS, retention, and engagement measurement – What’s working well and what needs immediate TLC.
The ability to make data-driven decisions – Even when user-level data is in short supply, while protecting your brand from the worst PR nightmares, and more importantly — protecting your consumers.

Bottom line is — brands are pouring money on their mobile apps but are unable to answer specific questions when it comes to their performance.

Questions like how certain campaigns ladder up to larger initiatives? How are these campaigns performing in certain regions? And how can we make our app experience better?

Marketers can’t fly blind when stewarding clients’ dollars.

Point #2 | The not-so-subtle art form of throwing money away

Because of lack of efficiency

Although mobile is the most data-rich marketing channel in existence, the sheer amount of media data that needs to be analyzed when it comes to assessing campaign performance — can be overwhelming and hard to gauge.

If you’re not working with an MMP, you’re probably spending countless resources and hours trying to make sense of way too many dashboards and spreadsheets. And unless you’re a computer, this manual analysis is a recipe for errors and plenty of missed opportunities for optimizing LTV and ROAS.

But when managed by a reliable, unbiased MMP, mobile measurement can pinpoint the value of specific channels, media sources, publishers, campaigns, and even creatives, while allowing you to focus on ensuring strategic growth.

Because of fraud

While we’re on the topic of wasting a precious budget, preventing ad fraud is one of the most effective ways to avoid that. That said, being able to prevent fraud effectively — calls for exhaustive market intelligence, scale, and advanced machine learning capabilities — which is why you need to partner with an established, mature MMP that holds a significant market share.

Buying your own fraudulent traffic and doing this over and over again on a growing scale will lead your business into the dreaded bleeding cash cycle.

To stay clear from the vicious, budget-draining fraud cycle, your MMP has to be able to remove bad players from the platform to ensure its integrity, and partner only with ad networks who take fraud prevention just as seriously.

Chapter 4

9 questions to ask an MMP before purchasing

Core functionality

1) Is your solution fully privacy compliant across all platforms?

The extremely limited access to IDFA after Apple’s release of the ATT framework, combined with Google’s upcoming deprecation of GAID, and the recent announcement of Apple’s privacy-centric SKAN 4.0 — all call for a completely different set of methods to be able to maintain mobile measurement.

Ask your potential MMP how they meet each of the above, and offer accurate, granular measurement while keeping your users’ privacy intact.

The MMP you choose should offer a compliance suite in their UI or accompanying API — that includes control and federation of access, deletion, and modification — to support CCPA and GDPR use cases.

Security wise, your MMP’s platform should meet the highest standards, including basic SSO, SSO enforcement controls, admin and role permissioning, JIT provisioning, audit logs, pen test report, SSL standards (and reporting), API, and uptime public reporting.

Last but certainly not least, be extremely wary of being linked with MMPs that violate privacy regulations, which unfortunately is more prevalent than what most would like to admit.

2) How comprehensive is your fraud protection?

Your MMP should be able to offer your business end-to-end fraud detection and prevention that includes coverage of real-time install fraud, post-install fraud, and in-app event fraud.

Ask your MMP if their SDK is closed source or open source — which is more susceptible to hacking.
Make sure they offer post attribution fraud detection. 16% of fraudulent activity can only be captured post attribution, so ensure your MMP has you fully covered.
What is your MMP’s market share? The larger it is, the more data they process, and the quicker it takes them to pick up on new fraudulent patterns.

3) How robust is your iOS14+ measurement?

Being able to tap into tangible insights in the age of user privacy is no walk in the park, which is why you need a top MMP player by your side.

Ask your MMP if, in addition to advanced privacy settings, they also offer:

A custom SKAdNetwork solution, which is not limited to predefined models.
Retroactive updates for post-install ATT consent.
Aggregated conversion modeling to account for null values from SKAdNetwork.
Advanced conversion value modes for revenue event funnels, and event frequencies that use up less SKAdNetwork Conversion Values capacity.

4) Does your solution include a data clean room?

Data clean rooms allow you to extract actionable insights in full compliance with user privacy standards.

Some refer to data clean rooms as “the Switzerland of data”, and rightfully so, because they offer a neutral, safe space for 1st-party user data to be leveraged collaboratively.

In a data clean room environment, two parties can securely share and analyze data with full control of how, where, and when that data can be used. While user-level data goes into the data clean room, aggregated insights come out in a co-mingled audience group (=cohort).

Ask your MMP if they offer a data clean room to allow access to user-level data without compromising your users’ privacy?

5) Does your solution enable advanced, cross-device, cross-channel attribution?

If you’re investing in multi-channel advertising, you have to be able to measure campaign performance holistically.

Ask your MMP if their solution includes:

Multi-touch attribution modeling – Scoring, weighing, and modeling multi-channels / platforms, as well as supporting integrations with multi-touch attribution providers / 3rd parties.
Cross-device and cross-channel attribution – Measuring marketing performance across mobile, desktop, CTV, OTT, and OOH.
CTV attribution – Attributing Connected TV campaign ads and QR codes.

Deep linking

6) Does your solution offer advanced deep linking capabilities?

Deep links are in essence customizable links that take users to any place on the web or a specific location in your app —- seamlessly and contextually.

Ask your MMP if their deep linking solution allows you to:

Customize various components such as the banner, metadata, and link domain or subdomain.
Create referral program links that can be attributed to the referring user or link (link aliasing).
Customize how links are displayed (especially in the context of viral referral campaigns or influencer programs on Instagram, Tik Tok, or any other social media platform).
Create web-to-app flows — such as highly customizable smart banners, customizable QR codes, and smart scripts.

Reporting & data visualization

7) How comprehensive and user-friendly is your dashboard UI?

If your MMP’s dashboard for campaign and performance management is not comprehensive and easy-to-use — you simply will not be able to get the quick insights or deep analysis you’re after.

Ask your MMP if their solution:

Enables segmentation across a wide variety of parameters — including time series, groupings by analytics, groupings by cohorts, platform, OS, OS version, DMA, etc.
Offers cohort analysis so you’re able to map user cohorts, constructs aggregate level reports, creates time-based analyses, and ingests standard campaign UTM parameters.
Offers raw data reporting on any attribution parameters (link clicks, impressions, installs, etc.) as well as custom events.

Integrations and APIs

8) How broad are your integrations with ad networks and major SRNs (Self Reporting Networks)?

Extensive measurement calls for extensive integrations.

Ask your MMP if they offer the following:

SKAdNetwork integrations – the ability to attribute SKAN campaigns across a large pool of relevant media partners, including a dedicated SKAN dashboard, and SKAN raw data functionality.

Bonus point: The ability to unify all SKAN and non-SKAN attribution data by deduplicating overlapping install data across the two datasets. Or in other words – a single source of truth.

Postbacks & postback APIs – including impression, view, view-through, click, install, reinstall & re-attribution elements, as well as an option to integrate directly to an API.
Customer Data Platform (CDP) / data management integration – the ability to integrate with a wide variety of CDPs, as well as major data hygiene and management providers.
Ad network integrations – the ability to fully integrate with the vast majority of ad networks.
Email service provider (ESP) integrations – the ability to integrate with all major ESPs.
Server-side and client-side API implementation options – clear documentation and support for integration with or without an SDK.
Integration with a wide range of SDKs – compatibility with standalone libraries on iOS, Android, and JS.
3rd party data integrations – the ability to integrate with key 3rd party platforms, including affiliate marketing tools and publishers.
Automated personalization – in-house models and 3rd party connectors to apply machine learning to datasets, and the ability to view or augment users with modeled characteristics.

Support, training, and customer care

9) What does your customer care, support, onboarding, and ongoing training include and what does your availability look like?

Your MMP needs to offer timely, attentive, knowledgeable, and cross-timezone support, onboarding, and training — to allow your business to fully adopt the platform and make the most of it.

One great way to learn if your MMP truly is customer-centric, is to ask about customers’ influence on the MMP’s product roadmap.

MMPs that take the time to listen to their customers, take their input into consideration, and actually apply it to their product — are the kind of MMP you want to partner with.

TL;DR

Here’s the gist of it all

The right MMP empowers you to make smarter, data-driven decisions, based on unbiased, indisputable data, so you can allocate your budgets, and optimize your app and campaigns’ performance, ROAS, and LTV.
The right MMP takes fraud very seriously, and invests heavily in multiple layers of fraud protection, across install, post-install, and event levels. Don’t waste your precious budget on scammers.
The right MMP allows you to run accurate, granular cross-channels measurement, so you can continually monitor and optimize your app’s monetization, user journey, retention, and user engagement levels.
The right MMP enables you to make swift, data-backed decisions even in the face of severe user-level data shortage, offering you advanced tools such as a single source of truth, privacy-centric data clean rooms, full SKAdNetwork support and Conversion Values studio, predictive modeling, and cohort analysis.
The right MMP allows you to stop paying double or triple for attribution. As you scale campaigns across several ad networks, let an impartial MMP connect all the dots and pinpoint exactly where credit is due.
Leave mobile measurement to the experts. The mobile landscape is a fragmented jungle, and a solid, experienced MMP is best equipped to have the resources, expertise, and cross-network scale to meet your demand for real-time, marketing insights you can actually trust.
Be sure to partner with a reliable, unbiased MMP with a secured and scalable platform that will do the heavy lifting for you, so you can focus on building a strategy and a loyal, higher LTV user base.