Attribution models explained

Zalify's five attribution models compared on one example order — Last Click, Last Non-Direct Click, First Click, Linear, and Any Click — plus how to choose.

An attribution model is the rule Zalify uses to split an order's revenue across the channels in its journey. You pick one from the dropdown at the top of the report, and it applies to every channel card and the Attributed column. This article walks all five models through the same example order so you can see exactly where they agree and where they don't.

One order, one journey

Take a single €200 order. Before checking out, the customer visited the store three times, from three different channels:

Meta (Paid)

They first discovered the store by clicking a Meta ad.

Google (Organic)

A few days later they searched and returned from Google (Organic).

Direct

Finally they came back Direct — typing the URL or using a bookmark — and placed the order.

So the journey is Meta (Paid) → Google (Organic) → Direct → purchase. Here's how each model splits the €200:

Model	Meta (Paid)	Google (Organic)	Direct
First Click	€200	—	—
Last Click	—	—	€200
Last Non-Direct Click	—	€200	—
Linear	€66.67	€66.67	€66.67
Any Click	€200	€200	€200

Same order, five different answers. The rest of this article explains each row.

Last Click

100% credit to the last channel clicked. In our example the last visit was Direct, so Direct earns the full €200.

This is Zalify's default and the simplest to reason about — it credits whatever was in front of the customer at the moment they bought. The trade-off is that Direct often wins, even though Direct isn't a channel you can market to; it usually means "the customer already knew you." That can make your acquisition channels look weaker than they are.

Last Non-Direct Click

100% credit to the last channel clicked, but Direct is skipped. Zalify walks back from the purchase, ignores any Direct visits, and credits the last marketing channel. In our example that's Google (Organic), which earns the €200 that Last Click gave to Direct.

When to prefer this over Last Click

Use Last Non-Direct Click when a lot of your orders close on Direct visits and you want credit to land on the channel that actually did the work of bringing the customer back. It answers "what was the last real touchpoint?" rather than "what was literally the last click?"

First Click

100% credit to the first channel clicked. Here that's Meta (Paid) — the ad that first brought the customer in — so Meta earns the €200.

First Click rewards discovery. It's the right lens for judging top-of-funnel and awareness spend: which channels introduce new customers, regardless of what closes the sale. Its blind spot is the mirror image of Last Click — it ignores everything that happened after the first visit.

Linear

Equal credit to every click. The journey has three visits, so each channel gets one-third of €200 — about €66.67 each.

Linear is the most balanced view: no single touch is over-rewarded, and every channel that participated shows up. Use it when your customers take multi-visit journeys and you want a fair picture of the whole path rather than crowning one winner. The trade-off is that a low-value touch counts the same as a decisive one.

Any Click

100% credit to each channel that appeared. Every channel in the journey gets the full €200 — so Meta, Google, and Direct each show €200.

Any Click credit is non-exclusive

Because each channel gets the full order value, the credited amounts add up to more than the order total — €600 across three channels for a €200 order. Any Click is not a way to split revenue; it answers a yes/no question: did this channel appear anywhere in the path to purchase? Don't sum Any Click revenue across channels or compare its totals against your real sales.

Any Click is useful for reach questions — "how often does Meta touch the journeys that convert, even when it isn't first or last?" — not for measuring each channel's exclusive contribution.

Choosing a model

There's no single correct model; each answers a different question. A quick guide:

If you want to know…	Use
What closed the sale	Last Click
What closed it, ignoring Direct returns	Last Non-Direct Click
What discovers new customers	First Click
A fair split across the whole journey	Linear
Whether a channel appears anywhere in the path	Any Click

Compare, don't commit

The fastest way to understand your channels is to switch models and watch the cards move. If a channel looks strong under First Click but weak under Last Click, it's an introducer — it starts journeys others finish. If it's the reverse, it's a closer. The gap between models is itself insight.

Next steps

Understand attribution in Zalify — the visit journey behind every order
Reading the attribution report — apply a model with the filters and columns

Understand attribution in Zalify

Reading the attribution report