Skip to content
AppDues
Log in
Contact salesSign up

Mojo: steering growth with ARPU and cohort reporting

Mojo is a Paris-based video editing app for short-form social content.

Its growth team treats average revenue per user as the compass for product, monetization and acquisition decisions. The original profile describes the team as having used a subscription management platform since launch and as continuing to treat it as the shared reference for revenue numbers. This guide walks through what the team measured, what it found, and how to copy the measurement pattern in your own app.

Why ARPU became the guiding metric

The growth team uses average revenue per user to judge whether changes to the product, the monetization model and user acquisition are working. By the end of 2023 the team reported a near doubling of ARPU year over year, expressed in its subscription analytics platform as realized lifetime value. Reliable, flexible revenue measurement is the precondition for that kind of claim: without a consistent definition and a consistent data source, a doubling is impossible to verify or to attribute. The same logic applies to any app weighing a similar result — decide first which revenue-per-user figure you mean, then make sure every dashboard and deck draws on that one figure before drawing conclusions from a change in it. Michal Zylak, Senior Growth Product Manager, is listed on the original profile; Michal Parizek, also Senior Growth Product Manager, is quoted on the day-to-day analytics work. Their roles are reported as given in the original material.

A measurement checklist you can reuse

The Mojo workflow translates into a short sequence any subscription app can follow:

  1. Pick one revenue-per-user definition and publish it to leadership so decks and dashboards agree.
  2. Cohort customers by acquisition date so later revenue can be tied back to the period and channel that produced them.
  3. Read realized lifetime value at fixed windows — for Mojo, day 14 matters because the trial runs seven days.
  4. Break metrics down by country and by conversion or lifetime window to expose segments that behave differently.
  5. Watch for sudden declines as a signal to investigate, not just a number to report.
  6. Feed the findings into pricing, paywall and discount experiments, then re-measure.

Each step depends on the one before it. A cohort without a stable revenue definition produces numbers nobody trusts; a country breakdown without a cohort frame mixes users acquired under different conditions.

Dashboard and chart use in practice

The product team monitors ARPU in its subscription analytics dashboard and applies custom filters to slice the data. Three uses stand out. First, cohorting inside the charts lets the team split results by different conversion and customer lifetime windows, so the same revenue figure can be examined on the horizon that matters for a given question. Second, the dashboard acts as an early-warning system. In the reported example, a regional paywall delivery failure showed up as a country-level ARPU drop, which the dashboard surfaced before it spread — a reminder that a country-level decline can be an infrastructure fault rather than a pricing signal. A useful check when you see one: confirm the purchase screen actually renders in that market before treating the drop as a price response. Third, location breakdowns support price testing. When conversion is weaker in some countries, that can indicate the price is set too high, which tells the team which localized experiments to run next.

Where revenue actually arrives: a discount-timing example

Data on what share of revenue lands at different points after install or onboarding changed how the team thought about discounting. Roughly half of new revenue arrived within the first 7 to 14 days, with the remainder spread from about three weeks out to one or two years. The team described this as a surprise and as a meaningful amount of revenue still to unlock.

Here is how that pattern can drive a decision. Suppose a cohort shows 50% of its revenue inside two weeks and a long tail afterwards. A discount offered on day 3 to lift early conversion mostly discounts revenue that would have arrived anyway, while a targeted offer aimed at the long tail addresses users who have not yet converted. The same cohort chart therefore answers two different questions — how much to discount, and when — and the answer depends on which window you read.

ConfigureTargetTest

How leadership uses the same numbers

The founders rely on the same platform for high-level figures — new and recurring revenue, churn and similar measures — and those numbers flow into the monthly decks shown to stakeholders and investors. That is the practical payoff of a single source of truth: the analyst's filtered chart and the investor deck draw on one definition, so a question about a number does not turn into a reconciliation exercise.

What comes next

The team frames its next phase around expansion into Asia, further work on mobile video editing and adaptation to generative AI, with the stated aim of making editing simpler and more intuitive. The growth plan rests on the measurement foundation described above rather than on a single tactic.

Owner-supplied details for this page

The following details are placeholders to be filled in by the page owner and are not asserted here. Each line is a blank worksheet entry rather than a published claim: Publisher name and contact email — to be supplied. Permission status for reproducing an existing customer story — to be supplied. Pricing, trial or plan terms referenced in calls to action — to be supplied. Current download, rating and ARPU figures, if they are to be restated — to be supplied. Date on which the quoted figures were last reviewed — to be supplied. Support or service commitments, if any are offered — to be supplied.

Mojo measurement workflow at a glance
StepWhat is measuredWhy it matters
Define the metricAverage revenue per user, expressed as realized lifetime valueOne definition keeps dashboards and leadership decks aligned
Cohort by acquisition dateUsers grouped by when they were acquiredTies later revenue back to the period and channel that produced it
Read realized LTV at fixed windowsDay 14, then extended to six monthsDay 14 reflects the seven-day trial; longer windows show the tail
Break down by countryConversion and revenue by locationReveals pricing opportunities and regional faults
Investigate declinesSudden ARPU drops in a segmentDistinguishes a broken paywall from a pricing problem
Test and re-measurePricing, packaging, paywall and discount experimentsCloses the loop between insight and action

Describes the workflow reported in the original profile; it is not a product specification and does not imply any particular feature set is available to other apps.

Mojo app profile (as published)
FieldValue
Websitemojo-app.com
CategorySocial video editing for Instagram, TikTok and similar platforms
HeadquartersParis
DownloadsOver 40 million worldwide
RatingsOver 12,000 ratings averaging 4.8 stars

These are the app's own published figures captured at an earlier date; they are not independently verified and should be checked against current sources before being restated.

Frequently asked questions

What does ARPU mean in this case study?

Average revenue per user — total revenue divided by users over a period. Mojo's growth team uses it to judge product, monetization and acquisition changes, and its subscription analytics platform reports the related figure as realized lifetime value.

Why does day 14 matter for Mojo's cohorts?

Because the app runs a seven-day free trial. Reading realized LTV at day 14 captures the point just after trial conversion, and the window can then be extended to six months to see how a cohort's revenue develops.

How does cohort analysis help when attribution data is limited?

Grouping users by acquisition date lets you observe what a group actually generated in revenue even when you cannot see which campaign produced each install. Combining that blended view with whatever attribution data exists gives a usable basis for budget decisions.

What is the difference between a broken paywall and a pricing problem?

A broken paywall is a delivery failure: the purchase screen does not appear for some users, so revenue falls for reasons unrelated to price. A pricing problem shows up as consistently weaker conversion in a market while the paywall works normally. A sudden country-level ARPU drop should be checked for the first before concluding the second.

Can this workflow be applied to a different app?

The sequence — one metric definition, acquisition-date cohorts, fixed LTV windows, country breakdowns, anomaly checks, then experiments — is general. The specific windows and thresholds depend on your trial length, price points and market mix.