Investor-ready metrics and cross-channel marketing ROI
Objective
Build a marketing analytics layer connecting Facebook, Google, Apple Search Ads, and OneSignal spend to hardware and software sales outcomes.
Obstacle
Marketing spend was distributed across four platforms with no unified attribution. Underlying revenue data required reconciliation against app store data.
Outcome
Delivered a full marketing performance layer across all paid channels, a validated investor dashboard with gross and net revenue.
Background
Carista is a leading provider of aftermarket car services, empowering drivers with advanced diagnostics, maintenance tracking and personalised vehicle insights.
We had already built a functioning analytics infrastructure covering subscriptions, app usage, Amazon hardware sales, and Shopify. But two significant gaps remained: the business had no way to measure marketing ROI across its paid channels, and it had no investor-grade view of aggregate performance.
With investor conversations becoming more formal, the pressure to produce credible, reconciled revenue figures and subscription metrics was real. At the same time, the marketing team was running campaigns across Meta, Google, Apple Search Ads, and OneSignal with limited visibility into what was actually driving conversions.
Challenges
No Attribution Consistency: Each marketing platform applied its own attribution logic. Facebook attributed conversions via cross-site cookies with view-through included. Google used a 90-day conversion window. Apple Search Ads had its own reporting structure. And UTM parameters in Facebook ad URLs were frequently malformed. The result was that platform-reported order numbers, spend, and ROAS figures couldn’t be compared or trusted. The marketing team had reports from multiple places that told contradictory stories.
Revenue Data Did Not Add Up: Carista’s revenue was being reported from Adapty, its subscription management platform. But when compared against Google Play and App Store figures, discrepancies of up to 13% emerged. Adapty was reporting higher revenue than the stores in some cases, due to missing tax deductions, differences in how refunds were handled, and data gaps on specific dates. Before any investor dashboard could be built, this needed to be investigated and resolved.
Solution
Unified Attribution Framework: We built a marketing attribution model using Google Analytics (GA4) visit data as the source of truth for click-based conversions, and applying a split between first and last paid touch before purchase. Platform-reported order numbers from Facebook and Google Ads were used for metric calculations like Cost per Order and ROAS, with UTM-attributed figures from GA4 retained as a parallel view.
Reconciliation Against First-Party Stores: We connected both App Store and Google Play data via Fivetran, then conducted a transaction-level reconciliation against Adapty. We adjusted our modelling to account for country-level tax deductions applied by Google Play and integrated Amazon Selling Partner commission data into both gross-to-net waterfall charts and the investor dashboard.
Impact
Marketing Spend With Measurable Returns: For the first time, Carista could see spend, ROAS, and attributed orders across Facebook, Google Ads, Apple Search Ads, and OneSignal in a single dashboard view. The team had a consistent attribution methodology to work from, and known discrepancies (like Facebook’s view-through counting versus Shopify’s last-touch model) were documented clearly so the marketing team could interpret numbers in context.
LTV Analysis: We conducted a deep dive analysis covering the UK, US, Germany, Spain, and Italy, examining hardware-to-subscription conversion rates by adapter type and protocol, iOS vs Android breakdown, cohort behaviour, and the time-to-conversion metric. This analysis identified patterns around secondhand adapter purchases inflating certain conversion rate figures, and separated true new subscriber acquisition from reactivations that were being miscounted in some Adapty definitions.
Investor Dashboard: Revenue figures in the investor dashboard were reconciled against first-party store data, with commission deductions and tax adjustments applied. The board-level view gave Carista’s founders and investment partners a clean, auditable picture of net revenue, subscription trends, and CLV , replacing informal reporting with a live, repeatable dashboard.
Creating Value For Carista...
We discovered over $150k of revenue being mis-reported from platforms,
Joined up attribution across 4 platforms ,
And gave their investors revenue numbers that reconciled against source data.
Get In Touch
Our friendly team are always on hand to answer questions, troubleshoot problems and point you in the right direction.
