PPC and Revenue Attribution: How to Prove What’s Really Driving Growth

If you’re running serious PPC campaigns, you’ve probably had some version of this conversation:

“We’ve spent £X on Google and LinkedIn this quarter. How much revenue did it generate?”

Not clicks. Not form fills. Not “brand lift”.

Actual pipeline and closed-won revenue.

Most marketing teams know PPC is pulling its weight, but proving it is a different story.

In fact, one 2025 report found that 41% of marketers say they can’t effectively measure marketing across channels, even though 63% rank ROI as their top metric. (Source: supermetrics.com) Another study found only 36% feel they can accurately measure ROI at all – and nearly half struggle specifically with multi-channel attribution. (Source: Firework.com)

So if your dashboards don’t quite match the story in your gut, you’re not alone.

This article walks through a practical, PPC-first view of revenue attribution for B2B companies:

• What revenue attribution really means (beyond last-click screenshots)
• The models that exist – and why they’re all imperfect
• How to wire PPC into your CRM so you can track deals and revenue properly
• How to combine attribution, experiments and broader modelling to get closer to causation
• Concrete steps you can take now to prove what’s really driving growth

And yes, we’ll keep it human. No mystical “data clouds”, no attribution wizardry. Just a framework your CFO, CMO and performance team can all live with.

1. Why PPC attribution is so hard right now

Multi-touch journeys, long cycles, and messy data

In B2B, the neat “search → click → demo → sale” journey is mostly fiction. The reality:

• Multiple people in the buying committee
• Dozens of anonymous touchpoints (search, comparison sites, community posts, review platforms)
• Weeks or months of lurking before anyone fills out a form

It’s no surprise that B2B marketers cite attribution as a top challenge, especially with long sales cycles and siloed data.

So when you log into Google Ads and see a tidy conversion number next to a campaign, there’s always a nagging question:

“Is that really what drove the deal, or just what happened last?”

Privacy changes have broken the old toys

Third-party cookie deprecation and tighter privacy rules are also eroding the old tracking methods:

• Removing third-party cookies threatens the accuracy of cross-site tracking and attribution modelling.
• Industry bodies have warned that cookie loss impacts “most data-driven capabilities” such as retargeting and measurement.
• Adobe research found that more than a third of marketers say cookie deprecation has already hurt their ability to track and measure engagement.

So we’re trying to answer harder questions… with weaker signals. Fun.

The trust gap

Despite the tooling, there’s a big confidence gap: only around a third of marketers believe they can measure ROI accurately, and nearly half say multi-channel ROI is especially difficult.

That’s why “proving PPC’s impact” can feel like performance theatre:

• You show last-click reports
• Sales say “but that lead came from a webinar”
• Finance sees conflicting numbers and cuts the budget “to be safe”

The job now isn’t to find a perfect model. It’s to build a robust, honest view of PPC’s role in driving revenue, and to communicate that clearly.

2. What “revenue attribution” actually means for PPC

Before we get into models and dashboards, it’s worth grounding on definitions. Most confusion starts here.

Revenue attribution vs. conversion tracking

• Conversion tracking = Did this click lead to any defined action? (e.g. form fill, signup, demo request)
• Revenue attribution = How much revenue (pipeline and closed-won) can reasonably be connected to that click or series of clicks?

The first is a pixel and a thank-you page.

The second is a data model plus CRM reality.

If your PPC reporting stops at “leads” or “MQLs”, you’re not doing revenue attribution yet – you’re doing lead attribution and hoping the conversion rate stays constant.

Sourced vs. influenced

In B2B, it’s helpful to separate two questions:

1. Sourced: Which channel created the first known opportunity or lead?
2. Influenced: Which channels played a meaningful role after that first touch?

Many leading B2B teams now track both sourced and influenced metrics rather than trying to cram everything into a single number.

For PPC, that means:

• Being realistic: search often “scoops up” demand that started elsewhere
• Still giving PPC credit when it re-engages accounts, moves deals forward or nurtures high-intent users

Conversion ≠ customer

Finally, one uncomfortable truth: not all conversions are equal.

• A £200 self-service signup and a £200k enterprise deal might look like “1 conversion” each in Google Ads
• A calendar booking from a student or a competitor technically “converts”, but doesn’t pay anyone’s salary

Revenue attribution is about going beyond raw conversion numbers and weighting PPC by:

• Deal value
• Segment/ICP fit
• Likelihood to close

That’s why wiring PPC into your CRM is non-negotiable (we’ll come to that).

3. The classic attribution models – and why they break in B2B

Most PPC platforms still push the same family of models. They’re useful lenses, as long as you know their blind spots.

Last-click attribution

What it does:

Gives 100% of credit to the last touchpoint before conversion.

Why people like it:

• Simple to explain
• Often matches how finance looks at “source”

Where it breaks:

• Ignores all earlier PPC (or non-PPC) interactions
• Over-credits branded search and retargeting
• Extremely biased in long, multi-touch B2B journeys

If your board only ever sees last-click, you’ll systematically under-invest in prospecting and upper-funnel activity.

First-click attribution

What it does:

Gives 100% of credit to the first recorded touchpoint.

Good for:

• Understanding which channels are genuinely creating net-new demand

Weaknesses:

• Ignores all the later work to nurture, educate and convert
• Can over-credit low-intent discovery channels
• Tracking can disappear through multi-device or a long time-to-conversion.

Useful for context, not for budget decisions on its own.

Linear, time-decay, position-based

These are all flavours of multi-touch attribution (MTA) – models that share credit across several touchpoints.

• Linear: equal credit to all touches
• Time-decay: more credit to touches closer to conversion
• Position-based (e.g. U-shaped, W-shaped): more credit to first and last (and sometimes mid-funnel) touches

MTA is widely accepted as a more realistic way to understand complex journeys, but it still has practical issues:

• Requires reasonably complete data (which cookies and privacy rules are now sabotaging)
• Still based on correlation – it can’t prove whether PPC caused the sale

Data-driven / algorithmic models

Platform “data-driven” attribution uses machine learning to assign credit based on patterns in your account. Helpful, but:

• It’s a black box – very hard to explain to finance
• It only sees what’s inside that platform (e.g. Google) – not your whole mix
• It still doesn’t automatically link to pipeline and revenue unless you’re importing that data back in

Takeaway:

Use attribution models as lenses, not gospel. None of them alone will “prove” revenue. To get closer to reality, you need to combine them with experiments and higher-level modelling.

4. Build a practical PPC revenue attribution framework

Let’s move from theory to something you can actually implement.

Step 1: Decide which revenue questions you’re trying to answer

A workable framework starts with a shortlist of questions like:

• Which PPC channels and campaigns are generating the most qualified pipeline?
• Which are driving closed-won revenue at a sustainable cost?
• Which play a strong assisting role (even if they rarely get last-click credit)?
• What is our payback period or LTV:CAC by channel?

If you can’t answer them today, that’s your requirements list for data and tracking.

Step 2: Clean up your conversion and event tracking

The foundation is still boring-but-crucial hygiene:

• Make sure every meaningful action is tracked once (not 12 slightly-different “form submits”)
• Align naming conventions for conversions across Google, LinkedIn, Meta, Microsoft, etc.
• Track key micro-conversions (content views, pricing page visits, product tours) as supporting signals – not primary optimisation goals for high-value B2B deals

Industry research consistently shows that lack of expertise and difficulty tracking touchpoints are top obstacles to effective attribution. Fixing that is the unglamorous bit that makes everything else work.

Step 3: Integrate PPC with your CRM

This is where you start doing revenue attribution, not just lead attribution.

Aim for:

• UTM discipline (or auto-tagging where possible) on every PPC campaign
• Lead and contact records that capture original source and campaign
• Opportunities / deals that are linked to those contacts
• Revenue fields (actual and forecast) that can be tied back to campaign and channel

At minimum, you want to be able to answer:

“For leads whose original source includes Google Ads / LinkedIn Ads / etc, what pipeline and revenue did we generate over the last X months?”

Once this is in place, you can start pushing CRM data back into ad platforms (offline conversion imports, enhanced conversions) so the algorithms optimise toward the right outcomes rather than just cheap clicks.

Step 4: Map your funnel and attach PPC roles

For each stage of your funnel, decide how PPC should contribute and how you’ll measure it:

• Top-of-funnel (problem-aware)
  • Goal: account engagement, audience building
  • Metrics: engaged visits, high-intent behaviours, new accounts added to CRM
• Mid-funnel (solution-aware)
  • Goal: drive qualified hand-raisers
  • Metrics: demo requests, trials, high-intent content downloads
• Bottom-funnel (purchase-ready)
  • Goal: convert in-market accounts
  • Metrics: SQLs, opportunities, closed-won revenue

Then label campaigns and ad groups by funnel stage. This stops you comparing a “discovery” LinkedIn campaign against a brand-only search campaign on cost per demo and calling one a failure.

5. Combine attribution, experiments and MMM (without needing a data science team)

Attribution models help answer “where should we give credit?”

To answer “what actually caused the revenue?”, you also need experiments and higher-level modelling.

Incrementality testing – proving lift, not just presence

Think with Google describes incrementality testing as the “gold standard” for understanding advertising’s true impact in a privacy-first way.

In practice, this looks like:

• Holding out a control group with no exposure to a given PPC campaign
• Comparing behaviour and revenue between exposed vs. non-exposed groups
• Measuring incremental lift in conversions, pipeline or revenue

You won’t run this for every campaign. But doing it periodically on key segments (e.g. brand search plus competitor terms, or your main LinkedIn programme) gives you a sanity check:

“If we turned this off, what would we truly lose?”

Marketing mix modelling (MMM) – the executive view

At the other end of the spectrum is MMM: a statistical model that estimates the contribution of different channels to overall business outcomes over time.

This used to be the exclusive domain of big brands with big data teams. That’s changing: Google has open-sourced their Meridian MMM framework to make modern mix modelling more accessible. blog.google

For a scaling B2B company, you don’t have to implement a full-blown MMM immediately. But you can:

• Start tracking consistent channel-level spend and revenue over time
• Incorporate external factors (seasonality, big launches, pricing changes) into your analysis
• Use even a simple regression-based view as a sense-check on what attribution reports are telling you

Google itself now recommends combining incrementality experiments, MMM and attribution to get a more holistic measurement picture.

The “good enough” stack for most B2B teams

If you don’t have a quant team on standby, a realistic stack looks like:

• Platform attribution (e.g. data-driven or time-decay) for day-to-day optimisation
• CRM-linked revenue reporting for channel and campaign performance
• Quarterly experiments on critical campaigns to understand incremental lift
• Simple mix analysis (even in a spreadsheet) using historical spend and revenue

Not perfect. But infinitely better than arguing over last-click screenshots.

6. Using your CRM to connect PPC to pipeline and revenue

Let’s zoom in on the CRM piece, because this is where most PPC revenue attribution lives or dies.

Minimum viable data model

At a bare minimum, you want:

• Contacts / leads
  • Original source (e.g. “Paid Search – Google”, “Paid Social – LinkedIn”)
  • First-touch campaign / keyword if possible
  • Key timestamps (created date, MQL date, SQL date)
• Accounts / companies
  • Linked to those contacts
  • Segment / ICP fit (industry, size, region, etc.)
• Opportunities / deals
  • Linked to account and all relevant contacts
  • Source and primary campaign
  • Stage, expected value and actual revenue

Now, for each channel (Google Ads, LinkedIn, Meta, Microsoft Ads), you can answer:

• How many new opportunities did PPC source?
• How many existing opportunities did PPC influence?
• What pipeline and revenue did those opportunities generate?
• What does that look like by campaign, keyword theme, or audience?

Avoiding the “all PPC or nothing” trap

One big tension in B2B attribution is that multiple channels want to “claim” the same deal.

A pragmatic approach:

• Use sourced revenue for budgeting and CAC calculations
• Use influenced metrics for strategy and storytelling (e.g. “PPC touched 60% of enterprise deals over £100k”)
• Be transparent with stakeholders about how you define each

The goal isn’t to win a political turf war. It’s to make better decisions about where to invest the next £10k.

7. Reporting that actually proves PPC’s revenue impact

Once the plumbing is in place, you can finally move beyond vanity metrics.

Dashboard views that matter

For B2B PPC, useful recurring views include:

1. Channel-to-revenue view
   • Spend → clicks → leads → opportunities → revenue
   • By platform (Google, LinkedIn, etc.)
2. Campaign-level revenue view
   • Pipeline and closed-won by campaign / ad group / audience
   • CAC and payback period where possible
3. Funnel velocity view
   • Time from PPC-sourced lead → opportunity → closed-won
   • Comparison vs. non-PPC leads
4. Segment view
   • Performance for ICP vs. non-ICP accounts
   • Performance by region, deal size band, or industry
5. Trend & cohort view
   • How PPC-sourced cohorts from prior months are maturing (to avoid punishing early-stage programmes that haven’t had time to convert yet)

Tell the story like an operator, not a vendor

When presenting, resist the urge to say “Google drove £X revenue” as if it happened in isolation.

Instead:

• Acknowledge where PPC is likely picking up demand created by other channels
• Highlight where PPC played a critical accelerating or rescuing role (“this campaign reactivated dormant opportunities and brought £Y back into the pipeline”)
• Be explicit about the model you’re using (e.g. “This view is last-touch on the opportunity; here’s the assisted-impact view for context”)

Trust comes from being honest about uncertainty and limitations – especially when budgets are under scrutiny.

8. Common attribution mistakes (and how to avoid them)

A few patterns we see again and again:

a. Optimising to the wrong conversion

If your main optimisation event is:

• A generic “ebook download”
• A “contact us” form that pulls in students, tyre-kickers and competitors
• A top-of-funnel event with no proven link to revenue

…you’ll end up with cheap “wins” that don’t make it onto the revenue board.

Fix: move towards optimisation on:

• High-intent actions (e.g. demo requests from ICP firms)
• Or, better still, imported offline conversions that represent qualified opportunities or revenue milestones.

b. Ignoring lag

Enterprise deals don’t close in 14 days because your attribution window does.

If you optimise based only on short-term conversion data:

• Early-stage demand gen campaigns will look “bad” and get cut
• Brand and search campaigns will look disproportionately “good”

Fix:

• Use longer lookback windows for analysis, even if your bidding uses shorter ones
• Introduce cohort reporting so you can see how older PPC-sourced leads mature over time

c. Comparing channels on the wrong basis

Comparing LinkedIn prospecting to Google brand search on cost-per-opportunity is like comparing a first date to a wedding on cost-per-minute.

Fix:

• Compare channels based on their intended role in the funnel
• Use blended views (e.g. “total CAC including upper-funnel spend”) to keep numbers honest

d. Falling for “perfect data” promises

Some tools sell the dream of flawless, deterministic attribution. In practice:

• Privacy rules and cookie limitations mean there will always be gaps
• B2B buying committees make it impossible to see every influence

Fix: aim for a “good enough” system that’s directionally accurate and trusted, rather than chasing an impossible 100% accuracy that never ships.

How Lever Digital approaches PPC and revenue attribution

At Lever Digital, we treat measurement as part of the strategy, not a bolt-on.

When we work with B2B teams on PPC strategy and revenue attribution, we typically:

• Start with the business model
  • Understand net margin, sales cycle, buying committee, ICP, and realistic payback windows
• Audit the current measurement stack
  • Tracking setup (GTM, GA4, platform pixels)
  • CRM data quality and lead/opportunity structure
  • Existing attribution models and reports
• Define the “minimum viable measurement” plan
  • Clean conversion events across platforms
  • Proper UTM structures and auto-tagging
  • CRM integration for leads, opportunities and revenue
• Build a mixed measurement framework
  • Platform attribution models for optimisation
  • CRM-based revenue reports for budgeting
  • Incrementality tests and simple mix analysis for strategic decisions
• Report in human language
  • Fewer screenshots, more “here’s what we tested, here’s what changed, here’s what we recommend next”

The outcome isn’t a magically perfect number – it’s a shared, credible view of how PPC contributes to growth, and a roadmap for where to invest next.

Want to prove what PPC is really doing for your revenue?

If you’re tired of hand-waving your way through board meetings with “modelled conversions” and conflicting numbers, you don’t need another vanity dashboard. You need a practical attribution setup that:

• Connects PPC all the way through to pipeline and revenue
• Balances day-to-day optimisation with strategic, incrementality-based decisions
• Gives your leadership team confidence that every pound of media spend has a clear job to do

That’s the work we do every day with B2B SaaS, FinTech and tech companies who care about profitable growth – not just more clicks.

If you’d like a clear, no-nonsense view of how to get there for your own setup, request a free proposal from Lever Digital.

We’ll review your current PPC, tracking and CRM data, outline where revenue is leaking (or being underreported), and suggest a tailored plan to start proving what’s really driving your growth.

‍