Lead Source Performance Scorecard: Compare the Same Customer Journey
November 19, 2026 Comment off
A useful lead source performance scorecard connects each purchased lead to its handling, customer outcome and cost. It compares groups with the same definitions and observation time, then states what the buyer should do next. A chart of lead volume and cost per lead is only the beginning.
Use this scorecard for an ongoing source review. For a new supplier, start with the lead-source pilot plan. For the daily workflow behind the numbers, use our purchased-lead management guide.
Choose the unit of comparison
Start with a defined group: a source product, delivery period and observation cutoff. Keep the product version visible. A supplier that delivers several inquiry types should not automatically receive one blended score. Separate meaningful differences such as product, delivery route or purchasing terms before explaining why results moved.
Choose the customer milestone you are measuring and define it in plain language. “Converted” could mean a booked appointment, a signed agreement or a completed purchase. Those events are not interchangeable. Keep event timestamps so a later reviewer can tell what was known at the cutoff.
Build the scorecard around questions
| Review question | Measure to show | Definition to preserve |
|---|---|---|
| What arrived? | Delivered and accepted records | Unique IDs and the acceptance rule |
| What was worked? | Assigned and handled records | The documented handling requirement |
| Who was reached? | Contacted customers | What counts as contact, counted once |
| What progressed? | Agreed intermediate milestones | Stage entry and event dates |
| What finished? | Customers reaching the final milestone | Outcome definition and observation age |
| What remains open? | Unresolved cases | No forced success or failure classification |
| What did it cost? | Lead charges, adjustments and handling | Consistent scope and allocation method |
| What happens next? | Decision, owner and review date | The evidence and limits behind the decision |
Show counts beside rates. Ten customers out of 100 accepted leads is a different quantity of evidence from one out of ten, even though both calculate to 10%. Neither rate alone explains uncertainty or whether the result will repeat.
Use denominators that answer the question
Contacted customers divided by accepted leads describes one part of the process. Customers divided by contacted customers describes another. Both can be useful, but switching between them without labels creates misleading comparisons. Do not remove unworked records simply to improve the headline rate; show the handling failure separately.
Use unique records for customer-level counts. Several contact attempts or several stage changes for one person should not become several customers. Resolve reporting duplicates using a documented rule while retaining the original delivery history needed for commercial reconciliation.
Compare groups at the same age
A group delivered six weeks ago has had more time to produce outcomes than one delivered last week. Compare both at a stated age, or show separate maturity columns. Keep a current operational snapshot too, but label it as a snapshot rather than a final outcome comparison.
Do not divide all sales completed this month by all leads received this month unless those are actually the same linked group. Older deliveries can generate current sales. Cohort reporting follows the original group through time and avoids silently mixing those populations.
Read a simple example without declaring a winner
Imagine two hypothetical groups, each with 100 accepted leads and the same observation age. A produced 40 contacted customers and eight completed purchases. B produced 50 contacted customers and eight purchases. Their accepted-lead-to-customer rates are both 8%. The contact rates differ, and the contacted-customer-to-purchase rates are 20% and 16%.
That pattern gives you questions, not proof of a superior source. Check handling, product mix and the definition of contact. If A cost $2,400 and B cost $2,800 in lead charges, their lead purchase cost per customer would be $300 and $350. Those figures exclude handling costs and do not measure profit. They are invented examples, not industry benchmarks.
Keep changes and adjustments visible
Annotate staffing gaps, offer changes, source changes and routing incidents. A stronger overall rate can come from receiving a larger share of an easier-to-convert product, even when each product’s rate is unchanged. Review the relevant subgroups before attributing a blended movement to source quality.
Separate requested credits from approved and applied adjustments. Use a consistent cost treatment and identify unresolved amounts. Then review the scorecard alongside your customer economics, including the cost of handling records that never became customers.
End with a decision someone can execute
- Repair: fix a documented delivery, assignment or reporting problem.
- Observe: allow immature outcomes to develop without pretending they are settled.
- Investigate: resolve a product-mix change or unexplained difference.
- Adjust: approve a bounded change in purchasing with a new review date.
Assign one owner and record the reason. “Source B is down” is a description. “Check the unassigned records, correct routing and review the affected group next week” is an operating decision.
Compare measurement practices at LGW
Bring the scorecard questions to conversations with potential partners. Review eligibility and apply for LGW’s $395 Lead Buyer Pass. Application and approval are required. Qualifying buyers purchase leads for their own use; lead sellers, publishers, agencies and call center providers are ineligible even when they buy leads.
