Are your leads any good? Build a lead-quality scorecard

Measure lead quality by following the same inquiries through to customers, not by celebrating a low cost per form submission. Start with one row per sales opportunity and clear rules for what counts as qualified, booked, and won.

You can do this in a spreadsheet before buying another tool. The useful question is: Where are potential customers getting stuck, and what evidence tells us why?

Decide what each stage means

For a local service business, these are workable starting definitions. Adapt them to your actual sales process before collecting results.

For example, you might define won as a first job completed and paid. A business selling longer projects might use a signed contract instead, while recording collected payments separately. Write the rule down. Do not switch definitions halfway through a comparison.

A repeat call about the same job updates the existing opportunity. Keep spam, tests, duplicates, and existing-customer requests identifiable so they do not silently inflate new-customer counts. Retain excluded counts to show how you cleaned the data.

Keep one record that survives the handoff

Use these columns in your private sales tracker:

Opportunity ID | Inquiry date | Source/campaign | Assigned person
Qualified date | Booked date | Won date | Current status
Loss reason | Next action/date | Collected revenue

Keep customer contact details in your access-controlled customer system. This scorecard can use an internal ID. Never put personal information in campaign URLs.

Use a short loss-reason list: outside service area, wrong service, timing, price, chose another provider, or unknown. Unanswered is not proof of poor lead quality. Record contact attempts and keep the reason unknown unless you have evidence.

Compare customers, not just cheap inquiries

Here is an invented comparison, not an Ocean Media client result or an industry benchmark. Both campaigns spend $1,000 on ads, use the same definitions, and have had equal, sufficient time for their leads to reach an outcome.

Campaign B has twice the inquiry cost but a lower advertising cost per customer. That does not establish profitability: customer value, fulfillment costs, fees, and other acquisition expenses still matter.

These small counts are a reason to investigate, not proof that B will keep winning. If no customers have closed, report the customer cost as not yet calculable, not $0.

Keep the denominators straight. A 50% qualified-to-customer rate is not a 50% inquiry-to-customer rate. When using our allowable lead-cost guide, match the close rate to the same lead definition used in your CPL.

Give leads time before judging the campaign

Group inquiries by when they arrived, then follow that group. Do not divide this week's sales from older inquiries by this week's new leads.

Choose a review age based on your sales cycle, and show how many opportunities remain open. If one campaign's leads are older, wait or compare equally aged groups. Keep later outcomes attached to the original inquiry date.

Then choose one investigation:

• Few inquiries qualify: inspect the actual requests, targeting, and service details.
• Qualified people do not book: inspect response ownership, availability, and objections.
• Bookings do not become customers: inspect cancellations and what customers expected.
• Outcomes are missing: repair the handoff before judging the advertising.

These are diagnostic starting points, not automatic explanations.

Make the dashboard labels match reality

Google Ads distinguishes qualified leads and converted leads. A converted lead reflects the conversion step you define; the label alone does not prove you collected payment.

Google Analytics lists separate recommended lead-generation events, including generating, qualifying, and closing a lead. Recommended events require implementation; a scorecard does not automatically connect your sales records to Google.

Before changing ad optimization, ask whoever manages tracking to trace one permitted test inquiry through the system and confirm what each event actually records. Keep diagnostic clicks separate from genuine inquiries and sales.

Start by reviewing one completed group of leads. Once its definitions and outcomes are reliable, test the economics with your own numbers.