Use case
Cutting the cost of a qualified lead
Lead volume looks fine, the pipeline is flat, and every channel reports success. The fix is usually not more traffic.
The situation
The dashboards look reasonable. Cost per lead is stable or improving, traffic is up, the agency reports are positive. Sales says the leads are poor. Pipeline has not moved for two quarters, and each side has data supporting its position.
This is one of the most common situations in B2B marketing and it is almost never solved by increasing spend. The cost per lead figure is measuring the wrong thing, and optimising against it has quietly taught every channel to find the cheapest possible enquiry.
What changes it is measuring against qualified pipeline instead, which requires a definition both functions agree on, tracking that survives consent loss, and a destination built for the traffic being bought.
How it shows up
- Marketing reports leads; sales reports revenue; the numbers do not reconcile.
- Cost per lead is falling and pipeline is flat.
- Ad platform conversion counts exceed actual enquiries.
- Campaigns point at the homepage.
- Nobody can say what a customer costs to acquire.
- Budget allocation between channels has not changed in a year.
Symptom, cause and change
The most expensive mistake in this situation is treating a symptom as a diagnosis. These are the three columns kept apart.
Why it happens
- Optimising to the wrong event
- Feeding form fills back to the platforms teaches them to find people who fill in forms, which is not the same as people who buy.
- No agreed qualification threshold
- Marketing and sales use different definitions, so both are right and the argument recurs monthly.
- Broken or partial tracking
- Consent loss and browser restrictions mean browser-side measurement observes a biased sample, and models built on it are confidently wrong.
- Unprepared destinations
- Traffic arrives at pages that were not written for the promise the ad made, which caps conversion regardless of media quality.
- Channel-level judgement
- Each channel is measured on its own metric, so none is accountable for the number the business cares about.
How we approach it
Audit the measurement first
Whether conversion tracking fires correctly, whether consent is handled properly, and whether the events being optimised toward are worth optimising toward. This regularly finds accounts that have been optimising to a broken signal for months.
Agree what qualified means
With sales in the room, in writing. This single artefact ends most of the recurring disagreement and is the prerequisite for everything else.
Connect the CRM back to the platforms
Offline conversion feeds so the optimiser targets qualified opportunities rather than form submissions. This is frequently the change that moves the number most.
Fix the destination
Dedicated landing pages per campaign promise, with the offer, the proof and the form aligned to what the ad said. Doubling a page’s conversion rate is cheaper and more durable than doubling its traffic.
Reallocate on contribution
Budget moved monthly between channels on contribution to qualified pipeline. Underperforming campaigns are cut rather than nursed.
Report with the uncertainty stated
Where consent loss means a share of conversions is unmeasurable, the report gives a range and names the assumption rather than a confident point estimate.
What changes
- One number both functions use
- Qualified pipeline from the CRM, which ends the argument about whose figures are real.
- Platforms optimise toward revenue
- Offline conversion feedback, so the optimiser finds buyers rather than form-fillers.
- Traffic lands somewhere built for it
- Dedicated pages per campaign, usually a larger conversion gain than anything inside the ad account.
- Budget moves
- Monthly reallocation on contribution, which is the single highest-return discipline available.
- Measurable acquisition cost
- Cost per qualified opportunity, traceable to a source you can open.
- Fewer, better leads
- Volume frequently falls while pipeline rises, which is the correct outcome and takes some explaining internally.
Where it goes wrong
The first difficulty is political. Reporting on qualified pipeline rather than lead volume makes the numbers look worse before they look better, and someone has to be willing to present a smaller number that means more. Programmes that cannot survive that conversation revert to vanity metrics.
The second is that the qualification definition gets set by marketing alone because sales is busy. That produces a threshold sales does not accept, and the argument continues with an extra document in it. The definition has to be genuinely joint or it is worthless.
The third is expecting the platforms to fix it. Bid strategy and audience settings are not where the problem is when the conversion signal is wrong. Changing them produces movement without improvement, which is worse than no change because it looks like progress.
A fourth is under-investing in landing pages while increasing media spend. It is the more expensive route to the same outcome, and it is chosen frequently because media is easier to buy than conversion work.
A fifth is false precision in attribution. Consent loss in Europe means a material share of conversions cannot be observed, and presenting a model’s confident output rather than a range produces decisions made on invented certainty.
Finally, impatience. Offline conversion feedback needs enough qualified events to train against, which in a long B2B cycle takes weeks. Reverting because the first fortnight looked flat wastes the setup work entirely.
Finally, there is a version of this problem that measurement cannot fix: the offer is weak. Where the proposition genuinely does not compete, better targeting produces better-qualified people who decline more quickly, and we would rather say that than spend a quarter optimising around it.
What else you could do instead
Before restructuring measurement, these are the cheaper things worth ruling out — and occasionally one of them is the whole answer.
- Fix the follow-up instead
- If leads are worked slowly or inconsistently, speed to first contact may be the actual constraint. It is one of the few reliably predictive variables and it costs nothing but process discipline.
- Change the offer, not the targeting
- A weak offer produces weak enquiries regardless of who sees it. Testing a different proposition is cheaper than restructuring the whole measurement stack.
- Raise the form friction deliberately
- Adding qualifying questions reduces volume and raises quality. Crude, unfashionable, and occasionally the fastest available improvement.
- Shift budget to search intent
- Where the market buys on active search, moving budget from interruption channels to search and SEO can improve qualification without any measurement work at all.
How we would know it worked
The primary measure is cost per qualified opportunity, read from the CRM rather than from ad platforms, with the baseline captured before anything changes. Platform-reported conversions stay in the report for in-platform optimisation and are explicitly not the number we manage against.
Alongside it we track the qualified rate — the share of enquiries that clear the agreed threshold — because that is the number that should move first. Total lead volume is reported for context and is expected to fall.
Where consent loss means a share of conversions is unobservable, we estimate the gap and state the assumption behind it, every time it appears. A confident point estimate built on a biased sample is worse than an honest range, and we would rather the report were duller.
How long it takes and what it costs
Measurement and tracking repair is typically two to four weeks. Landing pages and the qualification definition run alongside it. The first meaningful read on cost per qualified opportunity is usually eight to twelve weeks after the offline conversion feed goes live, because the platforms need enough qualified events to learn from.
In long B2B sales cycles the honest full read takes a quarter or more, because the qualified opportunities being counted have to work through to closure before the cost figure means anything.
Cost is a fixed monthly management fee against a defined scope, separate from media spend. We do not charge a percentage of spend, because it creates an incentive to recommend spending more — which is the opposite of what this situation needs.
Estimates are labelled as estimates. Timelines here are planning ranges from comparable work, not commitments, and not measured client outcomes. We quote against a defined scope after a discovery call.
Services involved
Digital Marketing
The layer above the channels: what the funnel is, which campaigns run, what each is meant to produce, and whether it did.
Read more →Paid Advertising
Buying attention at a price that works — with the tracking, creative volume and landing pages that decide whether it does.
Read more →CRM & Sales Systems
The system of record for revenue: where leads land, how they are routed, what happens next, and whether anyone can see the truth of the pipeline.
Read more →Website Design & Development
Sites built to convert and to survive — fast, accessible, integrated with your CRM, and maintainable by someone other than us.
Read more →Data Engineering & BI
Getting numbers out of the systems that hold them, into one place, in a state somebody can actually make a decision from.
Read more →Email, SMS & WhatsApp Marketing
The channels you own outright — where the list is yours, delivery is not rented from an algorithm, and automation does the follow-up nobody has time for.
Read more →Questions
Should we just spend more?
Almost certainly not, and it is the most common response to this situation. If the conversion signal is wrong and the destination is unprepared, additional spend buys more of the same unqualified enquiries at the same cost.
Why will lead volume fall?
Because optimising toward qualified opportunities rather than form fills teaches the platforms to find fewer, better-matched people. It is the correct outcome and it takes explaining internally, since the headline number gets smaller.
What if sales will not engage on the definition?
Then the project has a serious problem, and we would say so early. A qualification threshold set by marketing alone produces a document sales does not accept and the same argument with more paperwork.
How much does the landing page really matter?
Frequently more than anything inside the ad account. A dedicated page written for the campaign promise routinely outperforms a homepage by a wide margin, and it is cheaper to change than media budget.
Can you fix this without CRM access?
Only partially. Without the CRM there is nowhere for pipeline truth to live, and every measurement claim reduces to the ad platforms’ self-reports, which double-count across channels.
How long before the numbers improve?
Tracking repairs can show effects within weeks. The offline conversion feed needs enough qualified events to train against, so eight to twelve weeks is the realistic first read, and a full read in a long B2B cycle takes a quarter or more.
What about consent loss in Europe?
It is real and it means a share of conversions is unobservable. We implement server-side tracking and consent mode to recover what can be recovered, and we report a range rather than a false point estimate for the rest.
Do we need to change agencies?
Not necessarily. Frequently the existing supplier is optimising competently toward the wrong target. Changing what they are measured on can be more effective than changing who they are.
What does it cost?
A fixed monthly fee against a defined scope, separate from media spend, quoted after a discovery call.
What if this is not the right piece of work for us?
Then we say so, and it is a common enough outcome that it is worth stating plainly rather than treating as an awkward exception. The most useful thing a first conversation produces is frequently a recommendation to do something smaller, something different, or nothing at all.
Other situations
- Replacing spreadsheets with a real system
- Launching in a new European market
- Making company knowledge searchable
- Automating quote to invoice
- Recovering from a failed migration
- Inheriting undocumented software
- Passing a customer security review
- Merging systems after an acquisition
- Opening a second location
Recognise this?
Tell us what it looks like in your business. We will tell you what we would do about it, and whether it is worth doing.
Get in touch