Multi-touch attribution

Multi-touch attribution that sees the whole journey, not just the clicks.

Stella tracks every touchpoint behind a sale, asks the buyer what actually made them purchase, and joins the two on the same order. The path, the reason, and the margin.

  • Contribution margin, aMER and LTV:CAC, not a wall of impressions and click-through rates.
  • Attribution and post-purchase surveys in one tool, stitched together at the order.
  • Flat price from $750/mo. Your bill does not move when your volume does.
Book a strategy callStart at $750/mo

Included from Starter ($750/mo) and up.

Tracked pathSurvey answerHeard you on a podcastnever appeared in the pathFacebookOrganic SearchEmailDirectOrder #1047 · $186joined on order id

Same order, two independent records. The gold one is invisible to every pixel on the internet.

Trusted by growth teams who left enterprise measurement vendors

PopSockets
Momofuku
TUSHY
Rough Country
Neighbor
Pattern
Pirani
Plunge
WeightCare
Metro Vein Centers
Rylee + Cru
Miracle
PlantsBasically
The problem

Why does every ad platform claim the same sale?

Each platform grades its own homework and takes full credit for any sale it touched. Add the dashboards together and they claim more revenue than your store actually made. Multi-touch attribution exists to reconcile them into one view of one order.

Reported revenue, one month, one store
Meta Ads Manager$412,000
Google Ads$318,000
Email platform$196,000
Everything else$74,000
Claimed in total$1,000,000
Actually in the bank$640,000
The numbers that decide budget
Impressions 4.2M · Clicks 86.1K · CTR 2.05% · CPM $12.40
Contribution margin
$412K
real product costs, minus ad spend
aMER
2.14x
new-customer revenue per ad dollar
LTV:CAC
3.10x
by cohort, at 3 / 6 / 12 months

Clicks tell you an ad ran. These tell you whether the business is working.

Metrics that run a business

Nobody ever grew a brand off click-through rate.

Contribution margin from your real product costs. aMER, so retention cannot flatter your acquisition. LTV:CAC by cohort at 3, 6 and 12 months. Impressions and CPM are still there; they are just not the headline.

Where attribution fits

A click path is one third of the answer.

Attribution tells you what happened. It cannot tell you why, and it cannot tell you what would have happened anyway. Most stacks stop at the first question. Stella answers all three.

Three lenses on the same sale
Multi-Touch Attribution
what happened
What path did they take?

Every touchpoint the pixel can see, first ad to checkout.

Post-purchase surveys
what they say
What do they say drove it?

The buyer tells you directly, including the channels no pixel records: TV, podcast, a friend.

Holdouts & MMM
what it caused
What would not have happened anyway?

A controlled test. The only one of the three that proves cause, which is why the other two calibrate against it.

Each is a third of the answer. Stella answers all three.

How it works · Tracking

Your own domain, so the data keeps arriving.

Tracking runs on your own domain, built to outlast the browser privacy rules that cut most tracking off after about a week. Stella keeps recognizing a shopper for up to a year.

That is the shopper who clicked your ad, thought about it for three weeks, and came back. On most stacks that order arrives unattributed, and the channel that earned it gets cut.

How long a shopper stays recognized
Typical tracking cookie~7 days
Stellaup to 365 days
Runs onyour own domain
Tracking domainsup to 10
The same month, five ways
First touchMeta $286K
Last touchMeta $141K
LinearMeta $203K
Time decayMeta $172K
Position-basedMeta $228K
Spend did not change$94K
How it works · Models

See how much of your performance is a modelling choice.

Five models, first touch through position-based, each at a 7, 30, 60, 90, 180 or 365 day lookback. Switch between them on the same orders and watch the credit move. Nothing about the business changed; only the assumption did.

Every journey path opens into the orders behind it, so every claim is one click from the receipts.

One channel, three readings
TVYou know what you spent on it
––Multi-touch
Nothing. Not one touchpoint.
There was never anything to click.
2.4xSelf-reported
Buyers name it at checkout
Their answers against your spend, projected across total revenue.
1.9xIncremental
What it actually caused
A holdout or a calibrated media mix model settles it.

The first reading cannot see the channel at all, which is how good channels end up cut.

The channel your dashboard cannot see

Someone clicks Meta, searches Google, buys. The survey says TV.

The path credits Meta and Google, because those were the only things to click. TV never appears, so TV looks worthless, and eventually somebody cuts it. The same goes for podcasts, creators and a friend’s recommendation.

Stella asks the buyer instead, prices their answers against your spend, and lets a holdout or media mix model settle what it truly caused.

Where to start

Start where you are. The tools are already waiting.

You do not need geo holdouts on day one. What matters is that the next step is already in your account when you are ready.

Getting going

Attribution and surveys from $750/mo, with holdouts and media mix modeling included via upload. Most brands under $10M live here comfortably.

Around $8M to $12M

This is where being wrong gets expensive, and where holdouts start changing decisions instead of confirming them. Run them on the plan you already have.

Scaling seriously

Professional at $3,000/mo adds Always-On Incrementality, models running straight off your connected data, and a faster line to our team.

What else is in there

Built to be interrogated, not admired.

A funnel you can open

Impressions through to repeat purchases. Click a stage and get the actual orders behind it.

Customer LTV cohorts

Revenue per customer by acquisition month at 3, 6 and 12 month horizons, with blended CAC and LTV:CAC.

Period-over-period

Deltas that know which direction is good. Rising CPA is not a green arrow.

Ask it from your AI tools

On Professional and up, query your attribution and survey data from Claude, ChatGPT and the AI tools you already use.

Push margin back to platforms

On Scale, send conversions and contribution margin back to Meta, Google, TikTok and more, so their algorithms optimize on profit.

Not a black box

Everything is kept as raw, queryable rows in a dataset dedicated to your workspace. On Scale, delivered into your own warehouse.

What clients say

Hear it from the brands we run measurement for.

Jerel Blades
TUSHY

Multi-Touch Attribution is included from Starter ($750/mo) and up.

How to work with Stella

Have us run it, or run it yourself.

Most brands want measurement done for them. That's our consultancy, and it's where most clients start. Not ready for done-for-you? The same platform is self-serve, from $6,000 down to free. There's an option for everyone.

The consultancy · from $10,000/mo

We build your causal optimization loop, and run it for you.

Proper holdouts and MMMs are real data science, and most teams aren't staffed for it. So a PhD data scientist builds your loop, runs it end to end, and hands you decisions and board-ready proof you can defend.

Book a strategy callDone right, without dedicating the headcount.
A model built and fit to your data
  • A PhD data scientist runs it end to end
  • Holdout & region/cluster design (even dense, multi-location)
  • Results validated until they’re decision-grade
  • Board / CEO / CFO-ready decks
  • Active strategy + a testing roadmap
  • Dedicated Slack + a named team

$10k isn't for everyone. There's a plan for every budget.

Prefer to run it yourself? Same platform, self-operated: the full loop at $6,000, always-on incrementality from $3,000, attribution and surveys from $750, or start completely free. Walk down to the plan that fits.

Free
$0/mo
Run Stella’s real models on your own data.
Best for seeing what honest measurement looks like
  • Two media mix models and two holdout analyses every month
  • Upload a spreadsheet, get a real model back. No connectors to set up.
  • The same models we run for enterprise brands, not spreadsheet math
  • Ask questions about your results in plain English
Support: Not included
Start free
Starter
$750/mo
Know where your sales come from, and prove it with real tests.
Best for smaller brands doing $5M to $10M a year
  • Multi-Touch Attribution: see every touchpoint that leads to a sale
  • Ask buyers how they found you, right after they check out
  • Connect up to 5 sources (Shopify, Amazon, Meta, Google, TikTok) and stop pulling reports by hand
  • Run as many holdout tests and media mix models as you want, on data you upload
  • Ask questions about your results in plain English
Support: Ticket support
Start free
Most popular
Professional
$3,000/mo
Incrementality running every day, with no test to set up.
Best for teams who want measurement running all the time
  • Everything in Starter
  • Always-On Incrementality: a live view of how incremental every campaign is, updated daily, with no test to run
  • Holdouts and media mix models run straight off your connected data, region by region. Up to 7 sources, no more uploads.
  • See where the next dollar should go, and where a channel is already saturated
  • Ask your data questions from Claude, ChatGPT and your other AI tools
Support: Email support
Start free
Scale
$6,000/mo
Push what actually works back into your ad platforms.
Best for brands running many channels who want to act on the results
  • Everything in Professional
  • We send incremental conversions back to Meta, Google, TikTok and more, so their algorithms optimize for real profit instead of last click
  • All of your Stella data delivered into your own warehouse
  • Unlimited data sources
Support: Slack support and a named contact
Start free
FAQ

Attribution questions, answered.

What is multi-touch attribution?

Multi-touch attribution tracks every interaction a customer has with your marketing, from the first ad they see to checkout, and splits credit for the sale across those touchpoints instead of handing it all to the last click. It shows which channels start customer journeys, which close them, and what each is really contributing to revenue. It is an observational method: it maps what happened, and pairs with incrementality testing to prove what caused it.

Does my price go up as I get more traffic?

No. Stella is a flat monthly price starting at $750. It is not priced on your revenue, your order count or the number of events you send, so scaling your business does not scale your bill. Usage-priced tools quietly get more expensive in exactly the months you grow, which is the worst possible time to be rethinking your measurement stack.

What makes this different from the attribution tool I already have?

Two things. First, the metrics: alongside attributed revenue you get contribution margin from your real product costs, aMER on new-customer revenue, and LTV:CAC at 3, 6 and 12 months, instead of a wall of impressions and click-through rates. Second, post-purchase surveys are built into the same tool and joined to the same orders, so you can see what a buyer did and what they say made them buy.

Which attribution models do I get?

First touch, last touch, linear, time decay and position-based, each at a 7, 30, 60, 90, 180 or 365 day lookback, so you can see how much of your reported performance is a modelling choice rather than a fact. Survey answers come through as a sixth model, Self-Reported, in the same comparison.

How does tracking hold up on Safari and iOS?

Tracking runs first-party on your own domain, and Stella is built to outlast the browser privacy rules that cap typical advertising cookies at around seven days: a shopper stays recognized for up to a year. For someone who clicks an ad and buys three weeks later, that is the difference between an attributed order and an unattributed one.

Is multi-touch attribution the same as incrementality?

No, and anyone selling it that way is overreaching. Attribution shows you the path a customer took. It cannot tell you what would have happened if you had not run the ad. That takes a holdout test or a media mix model, both of which are included in every Stella plan, free and up. Attribution is how you see the journey; incrementality is how you prove the spend.

What plan includes Multi-Touch Attribution?

Multi-Touch Attribution is included from Starter ($750/mo) and up, along with post-purchase surveys. Holdout testing and media mix modeling are in every plan, free and up. Always-On Incrementality starts at Professional ($3,000/mo).

See the path and the reason on the same order.

Book a call and we will walk through what your own attribution looks like once the surveys are attached to it, or sign up and upgrade to Starter to run it today.

Book a strategy callor start at $750/mo