Skip to content

Multi-touch attribution

Attribution that shows why each order was credited

Compare nine models across configurable windows, from channel down to the individual creative — and open any order to see the touchpoints, click IDs and UTMs behind its credit.

  • First, last, last-evidenced, linear, time-decay, position-based, U, W and data-driven
  • Windows of 1 to 90 days, cash or accrual
  • Unknown stays unknown
Illustrative — sample data

Attribution capabilities

Nine models side by side

Rule-based models plus a data-driven Markov model. Switch models and see the change in each channel's credit without re-importing anything.

  • Data-driven shares are labelled estimated
  • Last-evidenced touch skips unknown sessions only when evidence exists

Windows and accounting modes

Lookback windows of 1, 7, 14, 30, 60 or 90 days, and cash or accrual accounting so revenue lands on the day of the order or the day of the click.

From channel to creative and product

Drill from channel to campaign, ad set, ad, creative and order, or pivot to products to see which ads sell what.

  • Keyword-level attribution for Google is not available

"Why was this order attributed?"

Every order opens an evidence panel: the store record, the customer, each touchpoint with its credit and evidence, a checklist of what was verified, and the lineage from payload to campaign.

Clicks plus platform views

Platform-reported view-through conversions can be added on top of click attribution, capped so they never exceed the unattributed and unknown share.

Customer journeys with honest gaps

A timeline of touchpoints, sessions, events and orders per customer, with explicit "gap — no data" markers instead of invented steps.

Know before you buy

What to know

  • Attribution needs the Deltryx pixel or server events to see sessions; history before installation cannot be reconstructed.
  • View-through data is only available in aggregate from the platforms, so it cannot be matched to individual orders.
  • Custom user-defined models are not available yet.

Questions, answered plainly

Which model should I use?

There is no single right model. Comparing a few side by side shows which channels depend on the rules and which hold up under all of them; tests and the mix model add an independent view.

What happens to orders with no touchpoint?

They stay in an "unattributed" row with unknown confidence. Deltryx does not redistribute them across channels.

Does it work for lead generation?

Yes. In workspaces without orders, lead and contact events are attributed as conversions rather than revenue.

See your own numbers, labelled honestly

Start a free trial and connect a store and an ad account, or book a walkthrough with our team.