#measurement

15 notes tagged “measurement”.

• Apple Ads measurement pitfalls — AdServices isolation, match rate, storefront country, CPP comparisonsThomas Petit's list of ways Apple Ads numbers mislead — AdServices attribution can't be compared with SKAN-measured channels, MMP-vs-Apple install "match rate" (80–90% is good), App Store storefront ≠ user location (up to ~30% non-US installs on US campaigns), and custom product page results that look 10× better only because they serve a different audience.
• Conversions API (CAPI) — why server-side conversion feedback mattersWhy browser pixels broke under Safari ITP and ATT, how server-to-server Conversion APIs restore feedback to ad platforms (identifiers to send, full event stream), which platforms now offer CAPIs, and why advertisers only integrate a few.
• Designing incrementality tests (duration, budget, CTV and brand)Practical rules for geo/holdout incrementality tests — run upper-funnel and CTV tests 4–6 weeks (two weeks shows nothing), give them enough budget not to be swamped, avoid launching into seasonality, and judge brand spend on a finance timeline.
• Do you need an MMP? AppsFlyer, Adjust and the state of SKAdNetwork (2026)Why mobile measurement partners (AppsFlyer, Adjust, Branch…) remain mandatory for app advertisers buying Meta at scale, what the 2026 AppsFlyer consortium investment means for neutrality, and why SKAdNetwork/AdAttributionKit is not worth heavy investment.
• Evaluating a marketing mix model (and whether Robyn/Meridian are biased)Assume any MMM is wrong until it passes three checks — parameter recovery on simulated data, out-of-sample forecasting of a period with real budget changes, and agreement with lift tests; open-source Robyn and Meridian are not rigged for Meta/Google, but most MMM vendors are thin wrappers around them.
• Google Play ASO specifics — listings, quality signals and data quirksHow Google Play differs from the App Store for ASO — description indexing and higher ratings bars, custom store listings (up to 50, Google Ads targeting), technical quality thresholds (ANR/crash) that cut browse traffic, app size, the Data safety form, and why Play "organic" and "search" numbers are mixed.
• How last-click attribution biases channel resultsLast-click flatters low-funnel, intent-capturing channels (search, Apple Ads, Google UAC standard) and under-credits view-based ones (YouTube, CTV, chatbot ads); incrementality tests have reversed channel rankings — and platform-reported numbers are biased toward the platform.
• How much organic does paid UA create? Measuring the paid-to-organic upliftThe routes by which paid spend lifts organic installs (rankings, browse/similar-apps, brand search), Playtika's regression-then-scale method for estimating organic installs per paid install (Brazil rewarded-video case 0.1 organic per paid via browse; France native-ads case ~0.4 via brand search), and why the answer depends on your organic mix.
• Knowing if your ecommerce brand is winning — CTC's three pillars (source of truth, shared target, execution)Luke Austin's operating framework for ecommerce growth — one trusted source of truth joining financial, marketing, order and calendar data; one shared definition of success (contribution margin at the top, daily targets laddering down to every channel); and relentless execution through calendar, ad, email and creative workflows plus MMM/geo-holdout testing — built in that order.
• Measuring the ROI of ASO and featuringHow in-house teams (Nordeus, Tilting Point, Zynga) measure ASO — installs as the north star, holistic vs single-activity ROI, a D7→D365 LTV multiplier when there's no predictive model, the featuring-uplift calculation, and using App Store Connect cohorts and unified tracking links.
• Meta's incremental attribution model — should you switch from standard?What Meta's ad-set-level "incremental" attribution model does (predicts which conversions the ad caused, and optimises for them) versus standard 7-day click / 1-day view / 1-day engaged-view, why results look worse after switching, and who should use it — Jon Loomer says most advertisers, Ben Heath says only larger advertisers with a customer base.
• Should we add a new ad channel? Scale, return, effort — and small platform syndromeHow to decide whether a new or smaller ad channel (CTV, AppLovin for e-commerce, chatbot ads, retail media) is worth testing — judge scale, achievable ROAS and team effort, require it to beat (not match) incumbents, and budget for a real test plus data integration.
• Signal engineering — auditing and shaping the conversion events ad platforms optimise onFirst check what conversion data actually reaches Meta/Google (duplicates, wrong optimisation event), keep "what the platform needs" separate from "what the business wants", and engineer enough purchase events (e.g. ~20/day per campaign in gaming) for campaigns to exit learning.
• The measurement stack — last-click attribution, incrementality and MMM togetherWhy app advertisers need both deterministic last-click attribution (for platform optimisation and daily decisions) and incrementality testing (for budget allocation), where MMM fits, and how to reconcile them against top-line results.
• When last-click attribution is good enough (and when to invest in incrementality)For small, performance-driven advertisers on one or two channels, last-touch attribution is often ~85% as good as experiments and better than a bad MMM; invest in incrementality tests as spend, channel count and brand effects grow.