MMP: How to choose a Mobile Measurement Partner (and why you need one)
A mobile measurement partner (MMP) is the instrument cluster in your mobile app growth journey. At a base level, MMPs provide mobile attribution service for app advertising campaigns. As the industry continues to mature, however, leading mobile measurement partners are evolving into cross-platform, full-function, globally useful marketing and advertising measurement platforms that offer instrumented visibility into growth campaigns.
The only question is whether your MMP is the altimeter in a single-seat 1937-era Piper Cub or a full modern guidance system in a SpaceX Falcon 9.
What do MMPs do: a short list
Growth marketers rely on MMPs for many, many factors:
- Measuring marketing results
- Optimizing in-progress campaigns
- Organizing campaign data
- Streaming data into BI systems easily, efficiently, immediately
- Enriching marketing and conversion data for value-added insights
- Calculating life-time value (LTV) of users, players, or customers
- Capturing and aggregating spend data across hundreds or thousands of partners
- Collecting data from major platforms like Facebook/Meta, Apple Search Ads, Google, Snap, Twitter, Tiktok, and other that only offer access to a select small number of partners
- Combining spend and conversion data to calculate overall and campaign-level return on ad spend
- Providing a single view of global performance
- Verifying ad partner performance
- Protecting advertisers from ad fraud
- Creating audiences for episode and live/dynamic marketing campaigns
- Navigating complexities in partner management by standardizing data definitions and terminologies, and normalizing both spend and conversion data
- Simplifying integration, compliance, and optimization for emerging mobile attribution models like SKAdNetwork and Privacy Sandbox for Android
- Providing multi-platform deep linking, deferred deep linking, and tracking/measuring link capabilities
- Ensuring legal compliance with all applicable privacy legislation and platform requirements
Ultimately, a mobile measurement partner is a growth tool advanced marketers use to collect, organize, understand, and activate all of their marketing data, particularly and especially with regard to mobile apps, but also on the web and other platforms.
For mobile, this means unifying all campaign data across paid and organic, social and search, including impressions and conversions, as well as app data including installs, opens, engagements, and key events. It means assigning marketing credit for installs and sign-ups and purchases, tracking of uninstalls, and more. For most brands, it also means cross-platform marketing measurement, web-based marketing analytics, plus other digital platforms. And for many, it even means integration of out-of-home advertising or email marketing or newsletter links so that marketing technologists can pull all the data from all their campaigns into one place.
The goals of all of this?
- Fully understand all marketing spend
- Completely grasp all conversions and results of marketing campaigns
- Knowing, not guessing, which marketing partners are driving the most success
- Real-time optimization of ongoing advertising campaigns
- Improved efficiency of marketing spend
- Unlimited scalability of advertising efforts
- Easy sharing of marketing activity to all stakeholders and teams
- … all of which sums up to faster, smarter, cheaper growth
Typically MMPs like Singular manage all this via an SDK in your apps, code for your websites, links for digital and even offline marketing, and server-to-server solutions for complete flexibility.
The big question: how to choose an MMP
In a sense, picking the mobile measurement partner that works for you isn’t massively different from picking any massive software platform. To make a decision, you have to accurately diagnose the need, decide what your goals are, build a team of stakeholders that care about the decision, and then move forward with the purchase process.
- Diagnose the need for an MMP
- Decide on your mobile measurement needs and goals
- Build a team of stakeholders including marketing, finance, BI, product, and executives
- Decide on budget
- Identify top options
- Shortlist the ones that look like potential fits
- Get a demo and feel out how the company responds to inquiries
- Get detailed specs and data from each of your shortlisted candidates
- Evaluate your options
- Make a decision
That will include identifying MMPs that are worth taking a look at (shortcut: it’s pretty much Singular, AppsFlyer, Adjust, Branch, or Kochava), deciding what your budget should be, and researching the vendors. In 2022, that often means a lot of searching, reading, and watching before actually contacting an account executive AKA sales rep — we get it — and asking colleagues and contacts for their opinions. (You can also check business and SaaS software review sites like G2 Crowd to see what other growth professionals say.)
Worth checking: Migrating MMPs? How to successfully launch with Singular
Then, probably after contacting your shortlisted vendors, getting detailed information from each of them, you’re going to evaluate your options, get a demo from the most promising potential partners, and get your boss or team to make a final decision and/or negotiate the contract.
An MMP, however, isn’t just a piece of software.
The best mobile measurement partners are platforms that integrate tightly with your data infrastructure as well as your marketing partners’ data infrastructure. They have teams that quickly respond to the always-fast-changing mobile advertising ecosystem. They have lightweight, capable, and configurable SDKs that ship inside your app or integrate via server-to-server technology, so they have to be completely robust, reliable, trustworthy, and transparent about what they do.
Most importantly, when you pick an MMP, you’re making a decision for years, not just today. That means you need to be 100% comfortable with the development pace and flexibility of the product team, 100% happy with the responsiveness of the customer success team, and 100% certain that any concerns or challenges that pop up in the future will be speedily and professionally resolved.
The even bigger question: what to look for in a mobile measurement partner
There’s some general functionality that you expect to get (but won’t necessarily) in every MMP, and there’s the specific product capability that you need and must have in your chosen partner.
In the big picture, your MMP needs to:
- Collect all your cost and revenue data, including from ad monetization
- Measure your performance
- Provide accurate attribution data (table stakes)
- Optimize your ad spend
- Prevent or alert you to potential fraud
- Fit well into your tech stack
- Connect to all the advertising platforms you do business with, or might potentially choose in the future (you don’t want future options limited by your MMP’s capabilities)
- Present insights in a way your team can understand and apply them
- Ship data where you need, when you need it, and in the specific formats that you need it
- Manage new technologies like SKAdNetwork or Privacy Sandbox for Android
- Be 100% transparent about data capture, storage, and use, so that you’re not put at risk as a result of unforeseen and unknown errors or liabilities
But of course, it gets much more detailed than that. And I’m going to dive into a lot more of those factors in a moment. But there are some soft skills and often-overlooked qualities that matter too, and I asked some of the customer support team at Singular what they might be.
Here’s what they shared:
What about reliability? Does the MMP you want to work with share uptime status and history? Can you be confident that your mobile measurement partner will be able to scale to meet peak demand, or could it crack under pressure?
What about customer support? Shockingly, not everyone instantly knows how to apply every tool. Nor do new processes and requirements forced by ecosystem changes like SKAdNetwork come naturally to everyone. Customer support that is fast, effective, and — particularly important — local and in your language matters.
What about number of partner integrations? As a fast-growing and nimble brand, you want the freedom to advertise anywhere you think you might find pockets of inexpensive high-value users, players, or customers. The last thing you need is to find out that your MMP doesn’t have a deep, official integration with either the industry’s leading platforms or an integration for the slightly-obscure-but-potentially-critical marketing partner you want to test. So: check the number of partner integrations and ensure it matches your expectations.
And ease of migration: most growing and big mobile brands have an MMP. The devil you know often seems better than the devil you don’t, and you know that switching will have costs, complexities, and perhaps a few tears along the way. Does the MMP you’re considering make it easy to switch? Do they do everything possible to make it simple? Will they walk with you and hold your hand during the process?
Some of these are challenging to quantify: talk to existing customers of the MMP you’re investigating and get the details from them.
Unsurprisingly, price also matters. Here the challenge is not so much to get the pricing details from each MMP but to dig deeper and find out how the pricing might change with volume, or what’s not included in the base price that you might really need, or whether you can get pricing that evolves with your needs.
Finally, and before we get into the detailed details, it’s important to get a sense of how innovative a partner you’re potentially bringing on board. Do they keep up with industry trends? Do they have the latest technology? Do they solve the newest problems? Are they always on the alert for changing conditions that require changing approaches to the core problems of attribution and marketing measurement?
The detailed questions to ask when picking a mobile measurement partners
OK. Now we’re getting into the nitty gritty: all the detailed details that you (sadly) must check before making a call on which MMP will be best for you.
(Note: there’s a full version of this with explanations and details that you can get. If you’re interested, book a meeting with Singular sales and they’ll be happy to walk you through it.)
Analytics: accurately collecting all your data
- What data granularities are supported for network-reported metrics (cost, impressions, clicks)?
- Will your solution be able to provide the following KPIs/Metrics: Cost (as registered on partners’ platform/dashboard), Impressions, Clicks, Installs, CPI, CPC, CTR, CVR, CPM, CVR?
- Besides cost data, what other dimensions and performance metrics are captured by your ad network integrations (for example: Network reported impressions, Creative Image, Bid amount, Campaign status, Video Views)?
- Describe methods utilized to fill in data gaps if a network is unable to report metrics by certain data breakdowns.
- Is your platform able to capture 100% of marketing spend across all platforms and devices, beyond mobile app advertising?
- Please detail spend collection methods utilized in your partner integrations.
- Does your solution standardize and normalize all the data pulled from networks? Please detail and provide documentation.
- How does your platform handle cases where network data doesn’t add up (ie, the sum of lower level granularities does not match the sum of higher levels granularities)?
- How are changes to historical data managed?
- Can you support cost reconciliations, retroactive data updates, discounts/credit from UA partners?
- How are statistics reported retroactively when Campaign Name is changed in the advertising platform? Detail how these are handled with any relevant documentation.
- Please provide a list of all supported partners for cost reporting.
- Is your platform able to ingest historical data from other third-party MMPs for trend analysis?
- When and how often is the data refreshed/pulled from partners?
- When is the latest data available? How often? How up-to-date?
Attribution: measuring marketing results and assigning credit for them
- Please explain your attribution methodology and waterfall.
- Is there support for multi-touch attribution?
- Support for media mix modeling?
- What is your fingerprinting methodology? What are your policies regarding privacy, iOS, and Android, and how will you ensure my app won’t be booted from the App Store for violations?
- What attribution configuration options are available?
- Are attribution configuration options determined via tracking links?
- What partner postback configuration options/customization are available?
- Is there support for partner event postbacks; event/attribution lookback windows?
- What tracking link parameters are available for source/network reporting?
- How do you approach uninstall/reinstall tracking? What is the deduplication logic/timeframe?
- How do you support retargeting/re-engagement attribution?
- How do you support deep linking and deferred deep linking?
- How do you support scalable tracking link management?
- How many partners are integrated with your MMP?
- What is the size of your SDK?
- Do you support custom user IDs supported for cross-platform (web & app) attribution and reporting?
- How long is an existing user recognized post-install? How long is an existing user recognized after their last session?
iOS and SKAdNetwork: enabling performance marketing even with limited data
- What is your approach to probabilistic attribution methodologies in iOS 14?
- What are the key elements of your SKAdNetwork solution?
- Please share mobile measurement partner documentation on your SKAdNetwork compatible SDKs.
- Which networks are supported by your SKAdNetwork solution? How do new networks get implemented?
- Do you manage the SKAdNetwork conversion value for the advertisers?
- What conversion models do you support? Can you combine multiple model types?
- Can you define business-specific revenue amounts for a Conversion Values model?
- What is your approach to cohort reporting in iOS via SKAN?
- How do you translate SKAdNetwork Campaign IDs to Ad Network Reported Campaign Names?
- How do you translate SKAdNetwork Conversion IDs back to internal/advertiser event names?
- Are you able to join cost with SKAdNetwork conversions for ROAS reporting at the app/source/country/campaign level?
- Describe the SKAdNetwork reporting and analytics tools you offer.
- Do you make raw SKAdNetwork postback data available? If so, what fields are included and how is it accessed?
Fraud detection and prevention: ensuring efficiency and effectiveness of ad spend
- What fraud detection mechanisms are available for synthesized impressions, clicks, installs, and/or IAPs?
- What fraud rejection mechanisms are available for synthesized impressions, clicks, installs, and/or IAPs?
- What toggles and customization for fraud prevention/detection mechanisms are available?
- What fraud reporting is available? Please include reporting granularity (ie fraud type, OS, app, country, source, campaign, publisher).
- What device-level fraud data do you make available? Are fraud postbacks supported?
Audiences: activating cohorts for remarketing/look-alikes/etc.
- What audience segmentation options do you support?
- How and where can the audience segments be distributed?
- How often are Audiences updated? How often are Audiences distributed & re-distributed?
Data access: automated ETL and API access to your data
- What is your availability and data retention policy lookback window for user-level log data (impressions, clicks, installs, events, postbacks, fraud)?
- Can we easily transfer and view ALL our data in our internal BI tool?. Does your solution allow for this? If so, how?
- What different options do you offer for accessing user-level log data (describe update frequency and customization options for each)?
- What API endpoints are available? What are the rate limits for each? Are there additional fees for API access?
Reporting: optimizing marketing efforts
- What reporting modules/functionalities do you offer?
- What metrics does the mobile measurement partner make available in reporting (cost data, DAU/MAU, retention, etc)?
- How often is data updated in each reporting module?
- What standard data dimensions are available for reporting marketing ROAS/CPA performance (ie. campaign, creative)? To what extent are reporting dimensions customizable?
- What are the options for report configuration and customization (ie reporting dimensions & metrics for each module)?
- How far back is data available in reporting? What are the reporting lookback windows?
- What cohort period breakdowns are available? What is the longest cohort period available? Is an LTV cohort available?
- Are custom-calculated metrics supported? (i.e. combined events, event conversion rate, ROI, “cost-per” event, “first-time” event); Please detail inclusion of custom metrics in reporting modules
- Can raw data be downloaded/made available outside of the platform programmatically?
- Does the platform have the capability to report cross-platform user flow (ie. track web spend that resulted in mobile app conversions)?
- Can you describe your capabilities for Campaign Performance Monitoring?
Implementation and migration: getting started
- Describe all implementation methods (ie SDK, S2S, CDP); include links for relevant documentation.
- What is your support level for a S2S only implementation?
- What features/functionalities vary depending on S2S implementation (vs. SDK implementation)?
- What fields/IDs are required for S2S installs/events?
- Do you support custom user IDs?
- Do you support real-time testing/debugging (SDK & S2S)?
- Do you support event timestamps/ordering, offline event caching?
- What support do you offer for migrating historical data from another attribution provider?
- Describe customer support capabilities (including time zones). Would we have a dedicated account representative?
Pricing and contracts: making it official
- What features/functionalities vary depending on price/package?
- Does pricing vary with contract length/commitment?
- Does pricing depend on installs, MAU/DAUs, custom events, or other parameters?
Summing up: picking a mobile measurement partner
It should be pretty obvious by now that selecting an MMP is hard. It’s not a process you complete overnight. And it takes the involvement of many people on your team.
The end result, however, is a data and measurement platform that works for you, that streamlines your workflow, and most importantly, that boosts your growth rate. Based on the massive growth in Singular clients over the past year and the huge number of new clients who have migrated from other mobile marketing measurement companies, Singular might just be that option for you.
Interested enough to get serious?
Book a meeting for a demo and chat today, and find out if it is.