Attribution platforms compared on published facts
Published facts let you compare platforms on method, what you install and what they charge, and they can't say which estimate is right. Each price below is dated, in alphabetical order, with no ranking.
€99 once, excl. VAT. Full refund within 30 days, no questions asked. You keep the read.
By Joris van Huët, Founder & CEOPublished 6 min read
Published facts let you compare platforms on method, install and price, and they can't tell you which estimate is right. Methods differ first, from touchpoint credit to geo experiments to media mix modelling. The prices differ next: Northbeam publishes $1,500 a month for Starter and $3,500 for Professional (northbeam.io/pricing, read 2026-09-08), while Haus publishes no price and sells four plans through a demo (haus.io/pricing, read 2026-09-26). The test a table can't replace is an experiment: a peer-reviewed comparison using 15 U.S. advertising experiments at Facebook found that observational methods often fail to reproduce the randomized result.
What is in the table, and what is left out?
The platforms are in alphabetical order, with no ranking. Every fact comes from this site's competitor registry: method and script are the registry's words, and each price is what the vendor's pricing page showed when the registry read it, on the date in the row. A platform with no recorded read date, or whose page could not be read, is left out, so absence says nothing about it. Prices change without notice; read the vendor's page again on the day you decide.
What do these platforms measure?
Google's ads measurement team, in an announcement dated 11 November 2025, split measurement into three families: MMMs give an overview of all media channels, sales and outside factors; incrementality gives causal data for specific campaigns and channels; attribution maps the path to purchase and assigns credit to touchpoints. Read each Method entry against those three. The announcement comes from a company that sells ads and isn't independently audited.
What do they publish about price and terms?
| Platform | Method | Script or pixel | Price as published | Terms and limits |
|---|---|---|---|---|
| AdBeacon | First-party cookie and click tracking | Yes | Starts at $299/mo, varying with monthly tracked revenue (adbeacon.com/pricing, read 2026-09-08) | Not in the registry |
| Admetrics | Bayesian statistics with multi-touch attribution | Yes | Growth €339/mo with €20,000 of ad spend included; Business €764/mo; Custom from €1,100/mo (admetrics.io/en/pricing, read 2026-09-08) | Annual billing is 15% less |
| Attribution App | Rules-based and algorithmic multi-touch attribution | Yes | Shopify plan from $199/mo; Pro from $399/mo (attributionapp.com/pricing, read 2026-09-08) | Optional setup of $2,500 to $5,000 |
| Billy Grace | Deep-learning multi-touch attribution with unified marketing measurement | Yes (own pixel) | €499/mo to €3,250/mo, plus Enterprise tiers (billygrace.com/pricing, read 2026-09-26) | Every tier is booked through a demo |
| Cometly | First-party pixel and server-side tracking | Yes (pixel, server-side) | Core from $750/mo, usage-based on monthly pageviews and quoted on a call (cometly.com/affiliates, read 2026-09-26) | Monthly or annual billing; no free trial |
| EYKData | Server-side multi-touch attribution on your own BigQuery | Yes (GTM, BigQuery, Stape) | Core €399/mo; Premium €529/mo (eykdata.com/pricing, read 2026-09-08) | Priced on gross sales over the last 12 months |
| Fospha | Impression-based MMM with multi-touch attribution | No | Lite $1,500/mo; Pro $2,000/mo plus a share of media spend; Enterprise quoted (fospha.com/pricing, read 2026-09-08) | Tiers are set by monthly media spend |
| Haus | Geo-lift experiments, with causal attribution and causal MMM calibrated on them | No pixel listed | No public price; four plans, each through a demo (haus.io/pricing, read 2026-09-26) | Basics runs geo tests in the US only; international testing starts at Core |
| Northbeam | Machine-learning multi-touch attribution | Yes | Starter $1,500/mo; Professional $3,500/mo; Growth and Enterprise quoted (northbeam.io/pricing, read 2026-09-08) | Starter is month-to-month; Professional and Enterprise are annual terms |
| Polar Analytics | Deterministic multi-touch attribution on its first-party pixel | Yes (Polar Pixel) | Core from $750/mo on the Shopify App Store, priced on online GMV (apps.shopify.com/polar-analytics, read 2026-09-26) | Discounts for annual terms; its pixel does not work retroactively |
| Prescient AI | Machine-learning marketing mix modelling | No | No public price; SMB, Growth and Enterprise tiers by business size and number of sales channels (prescientai.com/pricing, read 2026-09-26) | Annual contracts |
| Wicked Reports | Click-based multi-touch attribution | Yes (server-side, cart and CRM) | $499/mo, $699/mo and $999/mo tiers, by annual revenue band (wickedreports.com/pricing, read 2026-09-08) | Not in the registry |
The bases differ: AdBeacon prices on monthly tracked revenue, Polar Analytics on online GMV, Wicked Reports on annual revenue, EYKData on gross sales over the last 12 months, Fospha on monthly media spend, Cometly on monthly pageviews, Prescient AI on business size and number of sales channels. Read a published price as the tier you would land in, not as a quote for your store.
Where does a read on a GA4 export fit in the same columns?
One more entry in the same columns, for the read this site sells: what each channel caused next to what last-click gave it, with a data-health score and a next step for every channel.
| Platform | Method | Script or pixel | Price as published | Terms and limits |
|---|---|---|---|---|
| Causality Engine | Proprietary causal inference on a GA4 export | No; the input is the GA4 Attribution paths export as a CSV | €99 per read, one-time, excluding VAT; Pro €299/mo (causalityengine.ai/pricing, updated 2026-09-28) | Full refund within 30 days, no questions asked; works on GA4 channel groups, not campaigns, keywords, creatives or audiences; takes no ad spend, so it reports no ROAS |
How do I test a row before paying?
- Write your own numbers in the bases the table uses: monthly media spend, annual revenue, online GMV, monthly pageviews and number of sales channels. Each platform's price moves with one of them.
- Note which of your numbers sets each platform's price, then read its pricing page again today.
- Ask each vendor for the one thing the table can't show: a holdout, geo or lift test that its estimate for a channel was compared with.
- Pass: the vendor shows an experiment that agrees with its estimate in direction, or you run one on the channel where its credit and your last-click disagree most. Fail: the estimate can't be tested that way, so treat it as one more view, as you would any attribution rule.
Sources, 30 September 2026: Strengthen media measurement and ROI clarity with incrementality testing improvements (Google Ads Help, 11 November 2025); A Comparison of Approaches to Advertising Measurement: Evidence from Big Field Experiments at Facebook (Gordon, Zettelmeyer, Bhargava and Chapsky, Marketing Science, 2019); each vendor's pricing page as linked in the table, read on the date shown in its row and checked again on 30 September 2026.
Related answers
Frequently asked questions
Which attribution platform is cheapest?
Published facts can't rank them. Each platform prices on a different base (tracked revenue, ad spend, GMV, pageviews or plan tier) and some publish no price. Put your own numbers against each base, then read the vendor's page again on the day you decide.Do attribution platforms need a pixel on my store?
It depends on the platform. The table's Script or pixel column says yes or no for each one. Fospha, Haus and Prescient AI are listed as not needing one, and the GA4-export read at the end works from a file.How current are these prices?
Each row carries the date the registry read the vendor's pricing page, and every figure was found on the page again on the day this was written. Prices change without notice, so read the vendor's page on the day you decide.
Go deeper: Incrementality testing, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Keep reading
Terms in this article
- Attribution PlatformAttribution Platform is a software tool that connects marketing activities to customer actions. It tracks touchpoints across channels to measure campaign impact.
- Attribution SoftwareAttribution Software measures campaign impact by tracking customer interactions across touchpoints. It assigns value to each channel, showing what drives conversions.
- Causal AttributionCausal Attribution uses causal inference to determine which marketing touchpoints genuinely cause conversions, not just correlate with them.
- Causal InferenceCausal Inference determines the independent, actual effect of a phenomenon within a system, identifying true cause-and-effect relationships.
- First-Party CookieA First-Party Cookie is a cookie set by the website a user visits. These cookies provide essential website functionality, such as remembering user preferences and login information.
- IncrementalityIncrementality measures the true causal impact of a marketing campaign. It quantifies the additional conversions or revenue directly from that activity.
- Incrementality TestingIncrementality Testing measures the additional impact of a marketing campaign. It compares exposed and control groups to determine causal effect.
- Multi-Touch AttributionMulti-Touch Attribution assigns credit to multiple marketing touchpoints across the customer journey. It provides a comprehensive view of channel impact on conversions.