Quick answer
Last click attribution is a model that gives all of the credit for a conversion to the last touchpoint a customer clicked before converting, and none to any earlier interaction, so it shows which channel closed the sale but not which channels created the demand.
Formula: Last-click credit = 100% of the conversion to the final touchpoint; every earlier touchpoint gets 0.
What is last click attribution?
Last click attribution is a model that gives all of the credit for a conversion to the last touchpoint a customer clicked before converting, and none to any earlier interaction. It is simple and stable, and it answers one question well: which channel closed the sale. But it systematically undervalues the channels that created the demand, so if you use it, read it next to a first-click or linear view.
Mechanically, the model looks at each conversion path inside a lookback window, finds the final clicked touchpoint, and assigns the whole conversion to it. One path, one winner. A one-line example: Paid Social → Email → Paid Search → purchase gives Paid Search 1 conversion and the two earlier touches 0, even though they introduced the customer and brought them back. Attribution is the broader practice of deciding how credit is split across a journey; last click is the simplest rule in that family.
The rule became a default because it is easy to explain, easy to reproduce and needs no modeling: two analysts with the same path data get the same answer. Search Ads 360 still uses last click as its default model, and Google Ads keeps last click as a supported option alongside its data-driven default.
Last click vs last touch vs last non-direct click
Three names get used as if they were one model. They differ on two points: what counts as the winning touch, and how direct visits are treated.
| Model | Winning touch | How direct is treated | Where you meet it |
|---|---|---|---|
| Last click | The last touchpoint the customer clicked | A direct visit can win if it is the last click | Google Ads (last-clicked ad and keyword); SA360 default |
| Last touch | The last interaction of any kind, including impressions and direct visits | Direct visits and ad impressions can win | CM360 Last Interaction |
| Last non-direct click | The last click, skipping direct visits | Direct wins only when the whole path is direct | GA4 "Paid and organic last click" |
The split changes real reports. Take the path Display → Direct → purchase: last click and last touch credit Direct, while last non-direct credits Display, because the direct visit is treated as a return, not a source. GA4 states that paid and organic last click and last non-direct click are two names for the same model, and that all its models exclude direct unless the path consists entirely of direct visits.
CM360 adds one more wrinkle: its Floodlight model gives the full conversion value to the last click but falls back to the last impression when there was no click, while its Last Interaction model credits the last channel of any kind. Model availability is also dated: GA4 retired first click, linear, time decay and position-based in November 2023; its Attribution reports now offer data-driven, paid and organic last click, and Google paid channels last click.
The formula and a worked example
The formula is a counting rule: last-click credit for a channel = the number of conversions whose final touchpoint is that channel. Its share = credit ÷ total conversions. A revenue variant assigns the full conversion value to the final touchpoint instead of a count of 1.
Sample data: 3 purchases in a 30-day window, across three paths:
- Path A (4 sessions): Paid Social, day 1 → Organic Search, day 5 → Email, day 9 → Paid Search, day 12.
- Path B (2 sessions): Display, day 2 → Direct, day 10.
- Path C (2 sessions): Email, day 3 → Organic Search, day 7.
Credit under four models, in conversions:
| Channel | Last click | Last non-direct | First click | Linear |
|---|---|---|---|---|
| Paid Social | 0 | 0 | 1 | 0.25 |
| Organic Search | 1 | 1 | 0 | 0.75 |
| 0 | 0 | 1 | 0.75 | |
| Paid Search | 1 | 1 | 0 | 0.25 |
| Display | 0 | 1 | 1 | 0.5 |
| Direct | 1 | 0 | 0 | 0.5 |
| Total | 3 | 3 | 3 | 3 |
The arithmetic reconciles. Under last click, the winners are Paid Search (Path A), Direct (Path B) and Organic Search (Path C): 1 + 1 + 1 = 3. Under last non-direct, Direct's 1 moves to Display, and the column still sums to 3. Under first click, the openers win: Paid Social, Display and Email. Under linear attribution, each path splits evenly: Path A gives 1 ÷ 4 = 0.25 per touch, Paths B and C give 1 ÷ 2 = 0.5, so Organic Search collects 0.25 + 0.5 = 0.75. Every column sums to 3.
Now read the last-click column against the others. Paid Social, Email and Display appear across all three paths yet collect 0 last-click credit between them; Email alone assisted 2 of the 3 purchases. Direct gets 1 for a return visit that created no demand. First click flips the bias — the closers Paid Search and Organic Search drop to 0 — and linear shows who participated. The thesis in one table: last click is stable, but it systematically undervalues awareness channels.
The decision rule: a channel with 0 last-click credit but many assisted paths is not a cut candidate on this report alone. Check the first-click and linear columns before touching its budget.
Pros, cons and when last click is enough
Last click earns its place as a default. The question is when it is enough on its own.
Pros:
- Simple and reproducible. One rule, no modeling; anyone can recompute the winner from the raw path data.
- Low data requirements. It works with short histories and sparse paths, where data-driven models have too little to learn from.
- Sharp on the closing step. It tells you which channel, campaign or landing page actually finishes conversions — useful for checkout and landing-page work.
Cons:
- Zero for assists. Every channel that introduced or nurtured the customer gets 0, as the example above shows.
- Bias toward demand capture. Branded and paid search look stronger than they are; awareness channels look weaker than they are.
- Sensitive to definitions. The winner changes with the lookback window, direct-visit rules, click-vs-touch counting and broken cross-device identity.
When it is enough, as if-then rules:
- If most conversions happen in a single session, or the decision is about the closing step, then last click alone is enough.
- If you allocate budget across awareness and conversion channels, or journeys span weeks, then read last click next to first-click and linear columns; long B2B cycles are the extreme case — see B2B marketing attribution.
- If two tools report different conversion counts, then align the model, lookback window and direct/click rules before comparing numbers.
For the full menu of models — position-based, time decay, data-driven and when each fits — see multi-touch attribution; this page stays on last click.
Common mistakes
Four mistakes turn a stable model into bad decisions. All four come from reading the last-click column alone.
| Mistake | Why it misleads | Fix |
|---|---|---|
| Cutting assist channels on a last-click-only report | Email shows 0 last-click credit but 0.75 linear in the example | Check the first-click and linear columns before cutting; 0 last-click does not mean 0 contribution |
| Assuming GA4's "last click" counts direct | GA4's paid and organic last click is last non-direct; direct wins only all-direct paths | Read the model definition in the tool's official docs before comparing reports |
| Comparing conversion counts across tools | Model, lookback window and click-vs-touch rules differ, so the same journeys produce different winners | Align model, window and direct handling first, then compare |
| Reading last-click credit as causation | The model records sequence, not incrementality; the closer may have harvested demand created elsewhere | Treat it as a closing-stage report and test incrementality separately |
The fix is the same every time: never read the last-click column alone — pair it with a first or linear view.
How to read last click in TapCub
Run the three-path example above in TapCub. The point is not the last-click column on its own, but the last-click column next to the models that expose what it hides.
- Set the goal and select several models. In All-platform analytics › Behavior › Behavior analysis, choose attribution, set the purchase as the goal, and select last touch, first touch and linear together. Each model becomes its own column, so the Paid Social, Email and Display rows show 0 under last touch and non-zero credit right beside it — the assist channels are visible on one screen.
- Know which last-click variant you are reading. TapCub's touchpoints are the sessions before the conversion, and last touch gives the whole conversion to the final session, direct included. Path B therefore goes to Direct, as in the last-click column above. There is no last non-direct option, so read Display's role in Path B from the first-touch and linear columns.
- Choose what a touchpoint means. The dimension defaults to channel; switch it to UTM campaign or landing page to see which specific campaign or page closes the sale, not just which channel.
- Check the window and the closing rate. The lookback defaults to 30 days (1–90) and includes the converting session, so Path A's day-1 Paid Social touch, 11 days before the purchase, stays in. Each row also shows touchpoint conversion = paths ÷ touchpoint sessions in the range, which tells you how often a channel's sessions end up on a converting path.
Boundaries: TapCub credits sessions on your own site and apps; it is not an install-attribution provider, does no ad bidding, and scores up to 50,000 conversions per query. See compare attribution models for the full model list.
See also: What is an assisted conversion? Channels that help but never close.
Sources
- Google Analytics Help, Get started with attribution.
- Google Ads Help, About attribution models.
- Search Ads 360 Help, Types of basic attribution models.
- Campaign Manager 360 Help, About attribution modeling.
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Customer Experience Consultant
Writes about live chat, visitor conversations, and self-service, sharing practical ways to make customer communication clearer and more effective.