Attribution
that doesn't lie.
Last-touch attribution under-claims email and content, over-claims Meta and Google. Enter your channel spend + revenue and we'll show you what shared-attribution math looks like.
Last-touch is the single biggest cause of bad budget reallocation in mid-market marketing.
Marketing attribution is the question of which channels actually drove the customer to convert — and how much credit each deserves. Last-touch (the GA4 default for most accounts) hands 100% of credit to the final click before purchase, which is almost always branded search, direct, or a retargeting ad. The result: organic content + email lifecycle (which built the consideration) appear to drive nothing, while Meta + Google appear to drive everything. Budget then flows away from the channels that built the demand and into the channels that captured the click.
Shared attribution corrects this by spreading credit across the journey. The calculator above applies channel-specific factors based on patterns we have seen across hundreds of mid-market builds: email and organic typically deserve 25–40% more credit than last-touch awards; Meta and TikTok typically deserve 30–35% less. The output is directional — useful for budget conversations and sanity checks, not a substitute for a real measurement stack with server-side tracking, MTA platform, and incrementality tests.
- Under $25k/mo media: This calculator + GA4 + platform numbers is enough. Don\'t over-engineer.
- $25–100k/mo: Add server-side tracking (GTM SS + Meta CAPI + Enhanced Conversions). Restore 8–18% of attributed conversions.
- $100k+/mo: MTA platform (Triple Whale / Northbeam / Rockerbox) + quarterly incrementality (geo-holdout) + LTV modelling.
- $500k+/mo: MMM overlay alongside MTA. Quarterly portfolio reallocation. Channel mix decisions defended at board level.
Common attribution questions.
What is shared (multi-touch) attribution?+
Shared attribution is the umbrella term for any model that distributes credit for a conversion across multiple touchpoints in the customer journey, rather than awarding 100% to the last (or first) touch. Common models include linear (equal split), time-decay (more credit to recent touches), position-based (40/20/40 to first/middle/last), data-driven (algorithmically derived from converters vs non-converters), and Markov-chain (removal-effect simulation). This calculator applies a heuristic factor per channel: organic + email get a positive correction (last-touch under-claims them), Meta + Google get a slight negative correction (last-touch over-claims them).
Why does last-touch attribution lie?+
Last-touch awards 100% of conversion credit to whichever channel happened to be the final click before purchase — typically branded search, direct, or a Meta retargeting ad. Three structural problems follow: organic content + email lifecycle (which built the consideration) get zero credit; Meta + Google (which often catch the converted-already user) get inflated credit; and budget reallocation runs in the wrong direction (cutting the channels that drove the consideration to "scale" the channels that captured the click). Last-touch is the right model for zero analyses we have ever run.
When should I use this calculator vs a real attribution build?+
This calculator is a heuristic — useful for: (1) sanity-checking whether last-touch is materially distorting your channel decisions, (2) directional reallocation conversations with leadership, (3) early-stage pre-MTA brands. It is not appropriate for: deciding monthly budget at scale, defending CAC payback to investors, or replacing a real measurement stack. For brands above $50k/mo media spend, real attribution requires server-side tracking (GTM SS + CAPI + Enhanced Conversions), MTA platform (Triple Whale, Northbeam, Rockerbox), incrementality testing (geo-holdout or PSA), and LTV cohort modelling. We build that stack as a service.
What is the difference between MTA and MMM?+
MTA (Multi-Touch Attribution) tracks individual user journeys — works at the user level, requires deterministic identifiers (cookies, hashed email, server-side IDs), and is what platforms like Triple Whale + Northbeam optimise for. MMM (Marketing Mix Modelling) is a top-down statistical model — works at the aggregate spend-vs-revenue level, does not need user identifiers, and is increasingly the only viable measurement framework as iOS 14+ and cookie deprecation degrade MTA accuracy. Mature stacks run both: MTA for daily channel decisions, MMM for quarterly portfolio reallocation. Either alone is incomplete in 2026.
Why does iOS 14+ matter for attribution?+
iOS 14.5+ introduced ATT (App Tracking Transparency), requiring explicit user opt-in for cross-app tracking. Opt-in rates settled at 25–35% — meaning Meta, TikTok, and most paid-social platforms lost deterministic attribution for 65–75% of iOS conversions. Web tracking degraded similarly via ITP, Safari intelligent prevention, and the deprecation of third-party cookies. Server-side tracking (GTM Server-Side, Meta CAPI, TikTok Events API, Google Enhanced Conversions) restores 8–22% of attributed conversions on average — but never returns to pre-iOS14 deterministic accuracy. Probabilistic + MMM modelling fills the rest.
Can I trust the GA4 last-touch numbers I am pasting in?+
Partially. GA4 last-touch is roughly accurate for self-attribution (which channel got the final click in the user's GA4 session), but materially under-reports paid social by 11–18% post-iOS14 and double-counts cross-device conversions that share user IDs. If you have GA4 + a platform-side number (Meta in-platform, Google Ads in-platform), use the platform number for that channel's row in this calculator — it includes view-through and post-iOS14 modelled conversions GA4 cannot see. For organic + email, GA4 is the better source.