Measurement & optimisation
Measuring lead quality: from form to revenue
A measurement model that gives marketing, sales and suppliers a shared language for quality.
Section 01
Why CPL is not enough
Cost per lead shows what a form costs, not whether the person can be reached, fits the audience or becomes a customer. A cheaper campaign can create more work and less revenue.
A mature model treats the form as the start of the commercial chain. Each next step reveals whether improvement is needed in targeting, form design, validation, handover or sales follow-up.
Section 02
The quality ladder
Sales processes vary, but this ladder is a useful starting point:
Received: the enquiry was processed technically
Accepted: audience, location and required data match
Contacted: a real conversation occurred
Appointment or qualified call: need and suitability are confirmed
Proposal: a concrete opportunity exists
Sale: the enquiry created revenue and ultimately margin
Section 03
Connect events without personal data in analytics
You do not need names or email addresses in analytics to connect steps. A random reference ID can travel from landing page to lead, CRM and sale. Event IDs and page-view IDs prevent duplicate browser events.
Keep personal data in the operational system that needs it. Limit measurement to status, source, time, release and economic outcome to simplify analysis and reduce unnecessary privacy risk.
Section 04
Make feedback useful
Free-text labels such as bad lead cannot be analysed. Use standard reasons including outside region, duplicate, unreachable, wrong need, not an owner or timing too early.
Review patterns by source, campaign, vertical and release rather than debating every example. A rise in unreachable leads points to a different issue from a rise in wrong-audience leads.
Section 05
Compare equal maturity
New leads have not yet had the opportunity to convert. Compare cohorts at the same age, such as quality after 24 hours and sales after thirty days. Exclude test and preview traffic.
Set decision thresholds before results arrive: when to scale, investigate or stop. This prevents one unusually good or bad day from constantly changing direction.
FAQ
Frequently asked questions
01What is a good lead conversion rate?
There is no universal percentage. Compare with your own baseline by proposition, source and sales cycle, and assess margin rather than one intermediate rate.
02Does sales need to review every lead manually?
Not entirely. Technical rules can handle duplicates, missing fields and regions. Sales feedback remains necessary to understand intent and outcome.
03How do you avoid supplier disputes?
Agree definitions, validity criteria, rejection reasons and source data in advance. A shared factual model prevents anecdotes from controlling the relationship.