B2B Lead Attribution: Credit the Ads That Win Customers
B2B Lead Attribution: Credit the Ads That Win Customers

B2B lead attribution is the method you use to decide which marketing activity gets credit for a lead and, later, for the sale. For most Australian small and mid-sized businesses the answer is simple: record where every lead came from at the moment it arrives, carry that source into your CRM, and judge each campaign by the revenue it closes, not by the leads the ad platform reports. Fancy multi-touch models come later, if at all.
We run Google Ads and Meta Ads for Australian businesses, and the same problem turns up again and again. The platform report looks healthy, the sales team says the leads are poor, and nobody can say which one is right. Attribution is how you settle that argument.
TL;DR
- Lead attribution assigns credit to the ad clicks, visits and other touchpoints that led someone to become a lead, and then to a paying customer.
- Start with clean tracking and a CRM that stores the source of each lead. A model is only as good as the data behind it.
- Rule-based models (first touch, last touch, linear, time decay, U-shaped) are enough for most accounts. Data-driven models need conversion volume to work.
- Long B2B sales cycles need longer attribution windows, and the platform’s own number should always be checked against closed revenue in your CRM.
- A $99 AUD Google Ads audit shows where your current reporting and your sales disagree.
Table of Contents
- What B2B lead attribution actually measures
- Why platform numbers and your sales rarely match
- The attribution models, in plain terms
- What about multi-touch attribution?
- The tracking and CRM data you need first
- Set an attribution window that matches your sales cycle
- Where attribution goes wrong
- From attribution to a decision: a worked example
- What the reconciliation changes in the account
- Our view on attribution
- How we can help
- FAQ
- Sources
What B2B lead attribution actually measures
A B2B buyer rarely fills in a form on the first visit. They click an ad, read a page, come back through a branded search a fortnight later, talk to a colleague, then ask for a quote. Attribution puts a rule on that journey so you can say which touchpoints deserve credit.
There are three layers, and it helps to keep them apart:
- Lead attribution. Each form fill, call or enquiry is credited to a source. This is what most Google Ads and Meta Ads accounts report.
- Pipeline attribution. The lead is followed through the stages of your sales process, such as marketing qualified, sales qualified and opportunity. We cover that handoff in our MQL vs SQL playbook.
- Revenue attribution. Closed-won deal value is credited back to the campaigns and keywords that started or influenced the journey.
For a business paying for clicks, the third layer is the one that matters. A lead that never buys is a cost, whatever the dashboard says.
Why platform numbers and your sales rarely match
Google and Meta each count conversions inside their own system, and each is built to show its own channel in a good light. Two ads can both claim the same sale. Worse, some conversions are not new demand at all.
Here is a case from our own work. A real estate agency was told its cost per new client was $300. Its true cost was $800, because most of the reported conversions came from people searching the agency’s own name. Those people already knew the business. Once we separated brand searches from new demand and fixed the tracking, the cost per new client dropped to $68 within about three months and held for 10 months. We make no promise that your account will move like that, but it shows how far a platform figure can sit from the truth.
The gap is where the wasted spend hides. You only see it when leads are matched to what happened in the CRM afterwards.

The attribution models, in plain terms
Attribution models fall into two families. Single-touch models give 100% of the credit to one interaction, while multi-touch models spread credit across several. Here is what each does and where it fits.
| Model | How credit is assigned | Best suited to |
|---|---|---|
| First touch | All credit to the first interaction | Judging which channels create awareness |
| Last touch | All credit to the final interaction before the lead | Short sales cycles and simple Search accounts |
| Linear | Equal credit to every touchpoint | Teams that want a neutral view of a long journey |
| Time decay | More credit to touchpoints closer to conversion | Considered purchases with a clear decision point |
| U-shaped (position-based) | 40% to the first and last touchpoints, 20% shared across those in between | B2B journeys where first contact and conversion both matter |
| W-shaped | Extra weight on first touch, lead creation and opportunity creation | Businesses with a well-kept CRM and defined stages |
| Data-driven | Machine learning assigns credit from observed conversion paths | Accounts with enough conversions to train the model |
The W-shaped model only works if your CRM records when a lead was created and when it became an opportunity. If your team does not log those stages, the model has nothing to work with.
Which model should you start with?
Pick the simplest model that your data can support. Businesses without enough data infrastructure or conversion volume are better off starting with rule-based models, because a complex algorithm fed thin data produces confident but unreliable output.
A practical rule for the businesses we work with:
- Under about 30 conversions a month in an account: use last touch or a simple rule-based model inside the platform, and spend your effort on tracking and the CRM link.
- Steady volume and a sales cycle of several touchpoints: a U-shaped or time decay view is a sensible next step.
- High volume with clean conversion data: data-driven attribution can earn its place. Our guide to data-driven attribution covers when to trust it.
The 30 conversion figure is our working rule of thumb, not an industry standard. Treat it as a starting point and adjust to your own account.
What about multi-touch attribution?
Multi-touch attribution distributes credit across the customer journey instead of handing it to a single click. In theory that is fairer. In practice it needs every touchpoint to be recorded, and many are not. Conferences, phone calls, referrals and emails often leave no trace in your ad account. Attribution can also be complicated by offline-to-online interactions, such as conferences or direct mail, that lack event-level identity capture.
That is why a multi-touch report can look precise and still be wrong. If a third of your journey is invisible, the credit goes to the two-thirds you can see.
For a local trade, a law firm or a medical clinic, the journey is usually short enough that a clean last-touch view with branded searches separated out tells you most of what you need. For a multi-location business or a company selling higher-value services to other businesses, a multi-touch view starts to pay off, provided the CRM data is sound.
The tracking and CRM data you need first
No model rescues bad inputs. Before choosing one, check these five things.
- Conversion actions fire once and only on real enquiries. A tracking tag that counts a page view or a double form submission inflates every number that follows. If yours is unreliable, our conversion tracking fix page explains what we repair.
- The click ID is captured. Google adds a GCLID (Google click ID) to the URL when someone clicks an ad. Meta uses fbclid. Store these with the lead so the sale can be matched back to the click. Our article on closed-loop attribution shows how, including the privacy side under Australia’s APP 7 direct marketing principle.
- The CRM stores source, campaign and keyword. If the lead record only says “website”, you cannot tell Google from Meta from a referral.
- Sales stages are used consistently. Someone has to move deals to won or lost, with a value. Otherwise revenue attribution is guesswork.
- Won revenue flows back to the ad platform. Sending closed sales to Google Ads as offline conversions lets the bidding learn from customers, not just enquiries. See our guide to offline conversion tracking.
For revenue reporting you can also connect the CRM to accounting software such as Xero, so invoiced and paid amounts replace estimates.
Set an attribution window that matches your sales cycle
A window is how many days after a click the platform will still credit that click for a conversion. Defaults suit quick consumer purchases, not a B2B quote that takes months.
For Australian B2B sales, enterprise cycles often run six to twelve months, so a 90-day window misses the long-term effect of activity such as trade shows. The same source recommends extending the window to at least 180 days for complex, long-cycle sales. That guidance is aimed at event-led enterprise selling. A plumber or a dentist usually closes in days, so a short window is fine. Match the window to how long your own customers take from first click to signed job, which your CRM dates will tell you.
Where attribution goes wrong
Be aware of these limits so you read the results sensibly.
- Credit is not cause. A model shows who was present on the journey, not who made the buyer decide. Controlled tests, such as pausing a campaign in one region and comparing, give stronger evidence.
- Several people, one deal. In B2B, a number of stakeholders may research your business. If your CRM links only one contact to the opportunity, the others vanish from the report.
- Models built for simpler journeys. One study of over 700 B2B practitioners found nearly 90% still rely on single-touch or basic multi-touch models built for faster, simpler buying cycles. Using a basic model is normal. Knowing what it leaves out is the skill.
- Brand searches claiming credit. People who search your business name were usually already on their way. Counting them as new demand flatters the campaign.
The fix for each is the same habit: compare what the platform reports with what your CRM says was won, every month.
From attribution to a decision: a worked example
Say a professional services firm spends $6,000 a month on Google Ads and the platform reports 40 leads, a cost per lead of $150.
In the CRM you find that 14 of those leads were branded searches from existing contacts, 20 were genuine new enquiries and 6 were spam or wrong-fit. The 20 new enquiries produced 4 signed clients worth $5,000 each in first-year fees.
The true picture: $6,000 bought $20,000 of new revenue, and the cost per new client was $1,500. The $150 per lead on the dashboard told you nothing about that. The decision that follows is to cut the search terms that attract wrong-fit enquiries, protect the keywords behind the four clients and look at those terms first.

This example is illustrative and uses made-up figures to show the arithmetic. Your numbers will differ.
What the reconciliation changes in the account
Once leads are matched to revenue, the account can be improved in specific ways:
- Search terms. We review search terms weekly and add negatives so DIY and product searches stop eating budget. For a pest control company this, together with fixing tracking to capture GCLID and keyword data, moving from Maximize Clicks to Maximize Conversions and lifting the budget from $50 to $100 a day, brought the cost per lead down from $700 to about $100 in around six weeks.
- Match types. We use exact and phrase match with tight negatives. A plumber we moved off broad match keywords now sees around $3 to $4 back for every dollar spent, verified in his CRM.
- Budget. Money moves to the campaigns that close revenue and away from those that only generate form fills.
Results like these depend on the account, the market and the sales process, and we guarantee none of them.
Our view on attribution
We see the same thing in most accounts that come to us: the platform tells a good story, and the CRM tells a different one.
Our approach is to fix conversion tracking first, because every later number rests on it. Then we split brand searches from new demand so each lead is counted once, run weekly search term reviews with phrase and exact match, and check results against the sales in your CRM.
We do not think a business under a few hundred leads a month needs a sophisticated multi-touch tool. It needs one source of truth, kept honest, and a senior person reading it every week. That is why senior specialists run every account we manage, not juniors learning on your budget.
If you spend on Google Ads or Meta Ads and cannot say what a closed customer costs you, attribution is the first job, and it comes before any new campaign.
Jesse Lees
How we can help
If your reports and your sales disagree, start with a Google Ads audit for $99 AUD. We look at how conversions are tracked, which search terms are spending your budget and whether your numbers reflect new customers.
If you want us to run the campaigns, our Google Ads management starts from $1,725 AUD a month per platform, with no setup fee, no percentage of ad spend and a 3 month minimum. Meta Ads management is the same price, and our website tracking service covers the setup that attribution depends on.
FAQ
What is the difference between lead attribution and revenue attribution?
Lead attribution credits a source for each enquiry. Revenue attribution credits the closed sale value to the campaigns that helped win it. A campaign can look strong on the first and weak on the second, which is why we judge accounts on revenue wherever the CRM allows. Our piece on lead quality tracking shows how to measure leads through to sales.
Do I need a multi-touch attribution tool?
Usually not at the start. If your buyers decide within a few touches, a rule-based model plus CRM reconciliation gives you most of the value. Consider multi-touch reporting when your sales cycle runs for months, several people are involved in each decision and your CRM data is dependable.
What is the rule of 7 in B2B marketing?
It is a long-standing marketing rule of thumb that a buyer needs to encounter a business around seven times before acting. It is a guide, not a measured law. Its useful lesson for attribution is that one click rarely tells the whole story, so crediting only the last touch can undervalue earlier activity.
How long should I wait before judging a campaign?
That depends on how long your customers take from first click to signed job. Look at the dates in your CRM for your last 20 customers. If most closed within 30 days, review monthly. If they took three to six months, judge campaigns on a rolling view and use a longer window. Not sure where to begin? Our $99 AUD Google Ads audit is the quickest way to see how your account is counting leads today.
Sources
- Marketing attribution: what it is and how to use it (Adobe, Australia)
- Multi-touch attribution: what it is and best practices (Salesforce, Australia)
- Why 90 days is the wrong attribution window for B2B event ROI in Australia (B2B Insiders)
- Why teams can’t answer which campaigns created pipeline (LeadGenius)
Recommended
Keep reading
How we can help
Next step
Find out where your ad spend is going.
Get a $99 AUD Google Ads audit. We'll show you where money is being wasted and which leads you're missing out on.
Show me my current ad mistakes →