Cost per signed case: how to calculate it, and why cost per lead misleads
The one number that connects ad spend to new clients, how to count it correctly, and the mistakes that make it meaningless.
Most marketing reports for law firms end at leads. Leads are easy to count and they go up when budgets go up, so they make reports look good. But a firm doesn’t grow on leads. It grows on signed retainers. Cost per signed case is the number that connects the two: what you spent on advertising, divided by the clients that spending produced.
This guide covers how to calculate it, how to count it so it means something, and the mistakes that make it misleading.
Why cost per lead misleads
Two firms each spend $30,000 on ads in a month. The figures below are an illustration, not data from real firms.
| Metric | Firm A | Firm B |
|---|---|---|
| Ad spend | $30,000 | $30,000 |
| Leads | 400 | 250 |
| Cost per lead | $75 | $120 |
| Leads reached by a person | 220 (55%) | 212 (85%) |
| Signed cases | 15 | 24 |
| Cost per signed case | $2,000 | $1,250 |
On cost per lead, Firm A looks far better. On cost per signed case, Firm B is spending 37.5% less per client. The difference isn’t in the ads. It’s in what happened after the inquiry: Firm B reached most of its leads, and Firm A let almost half of them go cold.
A report that stops at leads would tell Firm A to keep doing what it’s doing, and might tell Firm B its ads are too expensive.
The formula
Cost per signed case = ad spend ÷ signed cases from the same lead-month cohort
Each part needs a definition, agreed before anyone reports on it:
- Ad spend: media cost paid to Google, Meta and other platforms. Agency fees are excluded here and shown separately.
- Signed case: a retainer signed by the client and accepted by the firm. Not a consultation booked, not a “likely” case.
- Same lead-month cohort: signed cases are credited to the month the lead first arrived, not the month the retainer was signed.
Count by lead month, not by signing month
Workers’ comp and injury cases often sign days or weeks after the first call. If you divide March’s ad spend by the cases signed in March, you’re mixing February’s leads, March’s leads and a few from January. The number moves for reasons that have nothing to do with March’s ads.
Counting by lead month fixes that. Here is an illustration for one month of leads:
| March leads, signed in | Signed cases |
|---|---|
| March | 12 |
| April | 6 |
| May | 2 |
| March cohort total | 20 |
With $30,000 of March ad spend, March’s cost per signed case is $1,500. Note the trade-off: the most recent month is always incomplete, because some of its leads haven’t signed yet. A good report says so, and updates the figure as cases come in.
Report attribution completeness next to it
Cost per signed case by channel depends on knowing where each signed case came from. In practice, some won’t have a known source: a referral who also saw an ad, a caller who dialed from a number written down weeks ago, a lead entered by hand without a source.
Attribution completeness is the share of signed cases with a known source. If it’s 70%, then cost per signed case by channel describes 70% of your cases, and the report should say that. If it drops, fix source capture before drawing conclusions from the channel numbers.
Show it with and without agency fees
Report cost per signed case twice:
- Ad spend only. This compares channels and campaigns fairly, because fees don’t usually change when you move budget between them.
- With agency fees included. This is what a client actually cost the firm in marketing, and it’s the number to compare against case value.
An agency that only shows you the first number is leaving out part of the cost.
Don’t compare practices as if cases were equal
A signed workers’ comp case and a signed trucking case are worth very different amounts. A cheap cost per signed case in one practice can be a worse investment than an expensive one in another. Report cost per signed case by practice and case type, and judge it against what those cases are typically worth to your firm.
What you need to measure it
You don’t need new software to start. You need four things connected:
- Source capture on every inquiry. UTM parameters, click IDs (gclid, fbclid), landing page and call-tracking source, saved into hidden form fields and your CRM.
- One record per person. Duplicates removed by phone and email, so someone who calls twice isn’t counted as two leads.
- The outcome in the same record. When a case signs, the CRM or case system marks it, with the date.
- A monthly export. Spend by channel from the ad platforms, joined to signed cases by lead month.
Most firms already have the first and the third in some form. The work is in connecting them.
Common mistakes
- Counting consultations as cases. Booked consultations are a step, not an outcome.
- Dividing by cases signed this month. See lead-month cohorts above.
- Blending practices. One number across workers’ comp, personal injury and employment hides more than it shows.
- Ignoring unreached leads. If 40% of leads never speak to a person, the problem isn’t the ads.
- Trusting platform-reported conversions as cases. Google and Meta report form fills and calls. Only your case system knows who signed.
What a useful monthly report shows
On the first line: cost per signed case, ad spend only and with fees, for each practice, by lead month. Then, underneath: leads, contact rate, speed to lead, consultations held, signed cases, attribution completeness and case mix. Every figure should say whether it’s verified (from your case system) or estimated (modeled).
If your current report doesn’t start there, ask for it. If the answer is that it can’t be done, that’s the first thing to fix. Our calculator will run the numbers for one month, and How we measure lists the definitions we use with every client.