Last-click attribution gives your paid search team credit for deals they didn't close. Here's what B2B SaaS companies get wrong about revenue attribution — and the model that actually works.
You spent $400K on marketing last year. Your attribution model says paid search generated 70% of closed revenue. Your CEO is thrilled. Your paid search team is getting budget increases.
But that model is wrong. And if you're making decisions based on it, you're probably cutting the channels that actually close deals.
This is the attribution problem. And if you're running last-click or first-click attribution in 2026, you have it.
B2B buyer journeys in 2026 average 6 to 8 touchpoints before a deal closes. Enterprise purchases often hit 10 or more. A single decision to buy $50K/year of software involves multiple people, multiple meetings, multiple content consumption events, and a demo — all before a contract gets signed.
Last-click attribution gives 100% of the credit for that $50K deal to the last thing that happened before the signature — usually a paid search ad. The thought leadership content your prospect read six weeks ago gets nothing. The referral that started the conversation gets nothing. The demo that sealed the deal gets nothing.
In one analysis of 50 closed deals, last-click attribution assigned 70% of closed revenue to paid search — even in deals where the sales team said the paid search ad was the referral source, not the close reason.
If you took that data and doubled your paid search budget, you'd be over-investing in top-of-funnel awareness while starving the channels that actually move deals through the pipeline.
The math doesn't lie, but the model might.
What it does: Gives 100% of revenue credit to the final touchpoint.
Why it's wrong for most B2B SaaS companies: It rewards the channel that happened to be present at the moment of close — not the channel that built the conviction to buy. A paid search ad at the bottom of a long consideration journey gets credited for work that content, demos, and sales calls did.
When it might work: Pure direct-response campaigns where the conversion happens in one session. Not relevant for B2B SaaS with multi-stakeholder, multi-session buying processes.
What it does: Gives 100% of credit to the first touchpoint in the journey.
Why it's wrong: It ignores everything that happened after the first impression. If your content team drove awareness but your sales team closed the deal, the content team gets credit and the sales team looks like a cost center.
When it might work: Brand-building campaigns where you need to justify top-of-funnel investment. Rarely useful as a standalone model.
What it does: Splits credit evenly across all touchpoints in the journey.
Why it's better than single-touch: It acknowledges that multiple interactions contribute. A 50-touch journey where every touch gets 2% credit isn't useful — but it's more honest than last-click.
The problem: It treats a LinkedIn cold outreach message the same as a product demo. Every touchpoint gets equal weight, which isn't how B2B buying works.
What it does: Touchpoints closer to close get more credit; earlier touches get diminishing weight.
Why it works better for longer sales cycles: It reflects reality — nurture sequences and late-stage interactions genuinely matter. In the same analysis that showed last-click crediting paid search at 70%, time-decay shifted substantial credit to mid-funnel content and product-trial activities.
The limitation: Still treats all late-stage touches equally. A demo and a contract renewal request don't have the same revenue impact.
What it does: 40% credit to the first touch, 40% to the last touch, 20% distributed across middle interactions.
Why it's the practical starting point for most B2B SaaS companies: It recognizes that acquisition and close both matter — and gives both their due. Middle touches still get credited, but not disproportionately.
How to use it: If you want to move from last-click to something better without a full analytics rebuild, U-shaped is your entry point.
What it does: Uses statistical models — Markov chains, Shapley value analysis — to estimate the incremental contribution of each channel across all journeys.
Why it produces the most accurate picture: It doesn't apply a predetermined rule — it looks at your actual data across all closed deals and estimates what each channel actually contributed, accounting for interactions and correlations.
The adoption gap: 41% of marketing organizations now use algorithmic attribution (2025 Salesforce data). The rest are running last-click on a $500K budget and calling it a strategy.
The implementation barrier: Requires a fully integrated Martech stack (CRM + ad platforms + product usage data). Most B2B SaaS companies aren't there yet.
When companies switch from last-click to multi-touch attribution, the findings are consistently uncomfortable:
Organizations implementing multi-touch attribution report an average 18–22% budget reallocation across channels and 12–19% reduction in CAC (McKinsey 2024). That's not a small number — that's the difference between profitable growth and a marketing team that burns cash.
You don't need a $200K analytics project to get a better attribution model. Here's a practical sequence:
Step 1: Move to U-shaped attribution in your existing tool. If you're on HubSpot, Salesforce, or any major CRM, you can configure position-based attribution in the reporting module. This alone will surface the gap between what last-click says and what's actually happening.
Step 2: Audit your touchpoint data. How many channels are actually sending data into your CRM? If your paid channels are sending signals but your content, email, and SDR activities aren't, your attribution model is only seeing half the picture. Fix the data gaps before you trust the model.
Step 3: Map attribution to revenue outcomes, not activity metrics. Tie each channel's attributed revenue to NRR, payback period, and LTV — not just closed ARR. A channel that closes low-ACV churn-prone customers has a different ROI than one that closes expansion revenue, even if the raw attributed ARR is the same.
Step 4: Reallocate 15% of budget and measure for 90 days. Pick the channel that multi-touch says has been under-credited (usually content or sales-assisted programs), move 15% of budget there, and run a 90-day test. If the channel attribution improves and CAC doesn't increase, you've got your proof.
Step 5: Move toward algorithmic attribution. When your data is clean and your stack is integrated (CRM + ad platforms + product usage signals), algorithmic attribution becomes the goal. Until then, U-shaped is the floor — not the ceiling.
If you want to know if your attribution model is broken, ask one question:
If this model told you to cut a channel entirely, would you have enough confidence in it to actually do it?
If the answer is no — if you'd second-guess the model because you've seen deals where the channel clearly mattered even though the model said it didn't — then you know your model isn't trustworthy.
Fix the model. Your budget decisions depend on it.
Last-click attribution is a historical artifact from a time when B2B buying was simpler and most deals closed in a single session. It's 2026. That's not how your customers buy anymore.
A B2B SaaS company running last-click attribution on a $500K marketing budget is making decisions about where to allocate six figures based on a model that systematically misrepresents reality.
The fix is not a six-month analytics project. It's moving to U-shaped attribution, cleaning your data, and reallocating 15% of budget based on what multi-touch actually shows.
Revcarto helps B2B SaaS companies build the revenue infrastructure that makes attribution actually trustworthy. If your marketing data and CRM aren't aligned, start with a free RevOps audit — we'll tell you what's actually broken.
Newsletter
No fluff. No templates. Just what we learn from doing this work.
Want this fixed?
We'll show you exactly where your revenue operations are breaking down — and what it'll take to fix them.
Get Your Free Audit →