Position-Based Attribution: The 40-20-40 U-Shaped Model

Position-based attribution splits a sale 40-20-40 — so on the $500 sale worked below, the first and last touches take $200 each and the two middle touches share the remaining $100.

Position-based attribution, also called U-shaped attribution, is the multi-touch model that gives 40% of a conversion’s credit to the first touchpoint, 40% to the last, and divides the remaining 20% evenly across the touches in between. It is the model for people who believe both discovery and conversion are the hard parts and the middle is nurture — a defensible claim, and one the flat linear split refuses to make.

It is popular in tools that ship it as a preset. TrackRev, deliberately, does not — it ships first-touch, last-touch, and linear, and this guide is honest about why.

First it works the maths so the 40-20-40 split is concrete, then it names when the U-shape genuinely fits, and how to reconstruct its insight from the models TrackRev does provide — each tied to real billing revenue, so the answer you act on is which channel pays your MRR.

Why this matters for your revenue

The two hardest, most expensive moments in most SaaS journeys sit at the ends: creating awareness where there was none, and converting an evaluated buyer into a paying one.

Single-touch models honour exactly one of those and zero out the other — first-touch credits only discovery, last-touch only the close.

Position-based is the model that refuses to choose, paying both ends 40% and treating the middle as the supporting nurture it often is.

For a funnel with a clear opening touch and a clear closing touch, that emphasis is closer to how the money is actually earned than either single-touch extreme or linear’s flat split.

The financial stakes are the same as with any model choice: budget follows the credited channel.

Awareness channels carry real lifetime value — across TrackRev workspaces, organic-search customers show a 2.1x LTV multiplier and direct a 2.3x multiplier (TrackRev platform data, Q2 2026) — yet they rarely deliver the final click, so last-touch alone underprices them.

Position-based restores the discovery end without erasing the close, which is its whole appeal.

But the 40-20-40 ratio is an opinion baked into a number, and if that opinion does not match your funnel — if the decisive moment is a mid-funnel demo, say — the model quietly misprices it.

Understanding exactly what the U-shape assumes is what lets you decide whether its answer is one you can defend. For the LTV data, see channel LTV per marketing source.

What position-based attribution is

Position-based attribution assigns fixed weights by a touch’s position in the journey: 40% to the first touch, 40% to the last, and the remaining 20% shared equally among all the middle touches. Plotted, the credit forms a U — high at both ends, low in the middle — which is where the “U-shaped” name comes from.

If a journey has four touches, the first and last take 40% each and the two middle touches split 20% into 10% apiece; scaled to a $500 sale, that is $200, $50, $50, $200.

The model encodes a specific claim: that the touch which created awareness and the touch which closed the sale did the heavy lifting, and the touches between them assisted.

That claim is often reasonable and sometimes wrong — a mid-funnel demo or webinar can be the decisive moment — which is why position-based is a considered opinion rather than a neutral measurement.

It is the weighted counterpart to linear: both credit every touch, but position-based bets that position matters and linear insists it does not.

Why the 40-20-40 split

The 40-20-40 weighting is a convention, not a law.

Its logic is that discovery and conversion are the two moments a marketer most wants to reward, so each gets the largest single share, while the assisting middle collectively gets a smaller fixed slice.

Other splits exist — some tools offer 30-40-30, which leans toward the middle, or 25-50-25 — and each moves money between channels without any change in customer behaviour.

That the ratio is adjustable at all is the first hint of position-based’s core weakness: the numbers come partly from a dial, not only from the data.

A worked example: a $500 sale

The same four-touch, 28-day, $500 journey used across this attribution series, so the U-shape sits comparably beside the other models. Position-based needs one small rule for the middle touches, but is otherwise simple to compute by hand.

The four-touch journey

Organic search brings the buyer in on day 1, a newsletter keeps them engaged on day 9, an affiliate review link pushes them along on day 21, and a paid-search ad brings them back to buy on day 28.

Organic is the first touch, paid search the last, and the newsletter and affiliate link are the two middle touches that will share the 20%.

How position-based credits the $500

The first and last touches take 40% each: organic search is credited $200 and paid search $200. The remaining 20% — $100 — is split evenly between the two middle touches, giving the newsletter $50 and the affiliate link $50.

So the split is $200, $50, $50, $200. Both ends are rewarded heavily; the two assisting touches each get a modest but non-zero share, which is exactly the outcome single-touch models cannot produce.

The maths, step by step

Assign 40% of the $500 to the first touch ($200) and 40% to the last ($200).

Take the remaining 20% ($100) and divide it equally among the middle touches: with two in the middle, each gets $50; with three, each would get $33.33; with one, it would take the whole $100.

The end weights stay fixed at 40% regardless of journey length — only the middle slice thins as more assisting touches appear. That fixed-ends, shared-middle rule is the entire model.

The same sale under all five models

The U-shape sits beside the other four models on the identical journey. It is the only row besides linear that credits all four touches, but it distributes them very differently — weighting the ends rather than spreading evenly.

Reading the comparison

Position-based and time-decay disagree most on the opener: position-based hands organic $200 because it was first, while time-decay hands it about $20 because it was oldest.

That is the two models encoding opposite beliefs — position-based says discovery is special, time-decay says recency is what counts.

Against linear the difference is subtler but just as real: both credit all four touches, but position-based pushes $200 to each end and only $50 to each middle touch, where linear insists on a flat $125 everywhere.

Which is closer to the truth is a claim about your funnel, not a fact the maths can settle.

The strengths and the blind spots

Position-based’s single design decision — reward the ends, share the middle — is both its strength and its liability.

What position-based gets right

Position-based is the only common single model that credits both ends of the journey heavily, and for a large class of SaaS funnels that is the honest emphasis.

Discovery is expensive and hard to defend under last-touch; the close is what actually banks the revenue.

A model that rewards both — while still giving the nurturing middle touches something rather than nothing — captures a truth that first-touch, last-touch, and even linear each miss in their own way.

When your typical journey really does have a clear discovery moment, a genuine period of nurture, and a distinct closing touch, the U-shape maps onto reality more faithfully than a flat split.

Where position-based misleads

The 40-20-40 split is an assumption hard-coded as a number, and it misleads wherever your funnel does not match it.

If the decisive moment is a mid-funnel product demo or a sales call, position-based lumps it into the thin 20% middle and systematically under-credits the touch that actually did the work.

It also carries a softer version of time-decay’s auditability problem: the weights are a tuning choice, so a channel’s credit can shift because someone picked 30-40-30 instead of 40-20-40, not because anything changed in the market.

And on short journeys it does no distinct work at all — with one or two touches it collapses to single-touch or linear, so it only earns its keep on genuinely multi-touch paths.

Who should use position-based

Position-based is a considered middle ground — more opinionated than linear, less absolute than single-touch. Match it to the shape of your funnel.

A good fit for two-ended funnels

Reach for position-based when your journeys have a clear discovery touch and a clear closing touch with genuine nurture between them — the classic case being content or a podcast that creates demand, an email sequence that nurtures, and a branded-search or retargeting click that closes.

Rewarding both ends is often the fairest reflection of where the work happened, and it lets you defend top-of-funnel spend and closing spend at once.

It is also a reasonable choice when you want a multi-touch view but find linear’s flatness too indiscriminate and time-decay’s recency bias too aggressive.

A poor fit for mid-heavy or short journeys

Avoid leaning on position-based when the decisive touch sits in the middle of your funnel — a demo, a trial-activation email, a webinar — because the model structurally under-weights it.

Avoid it too when journeys are typically one or two touches, where it adds nothing over linear or last-touch, and when you need a fully auditable model whose every number you can trace to a rule rather than a chosen ratio.

For long, many-touch B2B journeys where several touches genuinely matter, a flat linear view is often easier to defend; see attribution for long B2B sales cycles.

Position-based and TrackRev: an honest note

TrackRev does not ship position-based as a preset, and — as with time-decay — that is a deliberate design choice rather than a gap.

TrackRev ships first-touch, last-touch, and linear, and the U-shape is close enough to a blend of the first two that you can reconstruct its insight without adopting its fixed ratio.

Why TrackRev ships three auditable models instead

TrackRev’s three models are fully auditable: for any sale you can point to exactly why a channel got its credit — it was first, it was last, or it was one of N equal touches.

Position-based cannot make that promise as cleanly, because 40% of its answer rests on a 40-20-40 convention that is a choice, not a fact about the customer.

For most SaaS teams, an attribution figure you can explain to finance and defend in a budget meeting is worth more than a weighted split that is marginally more flattering to your instincts but harder to justify.

The reasoning is the same one that keeps time-decay out — explainability over a tuned parameter — set out in the attribution-models comparison.

Explainability over a fixed split

The objection to position-based is auditability, in a gentler form than time-decay’s.

When a channel’s position-based credit changes, you often cannot tell whether behaviour shifted or the split simply reweighted it — and at the conversion volumes most SaaS businesses have, a small-data model behaves more reliably when it has no tuning dial to obscure the signal.

A model whose weights are 40-20-40 by convention is a weak foundation for moving budget; a model whose rule is “first”, “last”, or “equal” is not.

Approximating the U-shape with first and last touch

You can get most of what position-based tells you from the models TrackRev ships, because the U-shape is essentially first-touch and last-touch blended, with a thin middle.

Read first-touch to see the 40% discovery end — which channels create demand — and last-touch to see the 40% conversion end — which channels close.

The two ends the U-shape emphasises are exactly the two single-touch views TrackRev provides, and running them side by side surfaces both explicitly, without committing to a fixed 40-20-40 ratio you would then have to defend.

Add linear for the middle, and you have bracketed everything position-based would have told you. See multi-touch attribution for SaaS for how the shared click log makes switching instant.

When NOT to use TrackRev

If a position-based or time-decay preset is a hard requirement for your reporting, or you need a black-box data-driven model trained on a large dataset, TrackRev is not the tool — it stands behind three auditable models on purpose.

If your billing is not on Stripe, Paddle, Polar, or Lemon Squeezy, the revenue join does not apply, and if you sell physical products or run a CRM-led enterprise motion, a specialist tool fits better.

TrackRev is for SaaS teams that would rather have attribution they can explain than a weighted split they cannot.

The stack maths is the consolidation case: a Bitly Growth plan (~$35/mo) plus a Rewardful Starter plan (~$49/mo) is $84+/month for two tools with two definitions of a sale, while TrackRev is $39/mo for link tracking, attribution, and the full affiliate programme on one.

See the revenue-attribution product for the connections.