Judge your ads on a full quarter of signed cases
Published Nov 13, 2026 · Last checked Nov 13, 2026 · Last changed Nov 13, 2026Large retailers ran 25 ad tests, and most of the tests could not tell a profit from a loss. A weekly ad report tells a law firm even less. Judge each channel on a full quarter of signed cases, and test with one large change.
- Verdict
- ACT
- For
- All practice areas that buy ads. Firms with 2 to 50 lawyers. Owner, marketing.
- Cost and time
- No cost. The test saves the channel's spend for four weeks. About two hours to pull the numbers.
Why
Researchers studied 25 online ad experiments run by US retailers. The ads cost $2.8 million in total. Most experiments covered more than a million shoppers. Only 3 of the 25 could tell a campaign with no return from one with a 50% return. Sales from one shopper to the next varied too much. A law firm has far fewer cases than a retailer has sales. So a small move in a weekly report says nothing. A full quarter and one large change show much more.
What to do
- Stop judging a channel by its weekly report.
- For each channel, write down the signed cases and the fees from the last full quarter.
- Write down what the channel cost in the same quarter, with the agency fee.
- Pick the channel you doubt most. Turn it fully off for four weeks. Do not trim it by a small amount.
- Count new inquiries, booked consults and signed cases from all sources in those four weeks.
- Compare with the four weeks before. If the numbers held, keep the channel off for the quarter and move the money.
- Ask your agency to report signed cases, and not clicks.
The rule to check
Read your agency contract before you pause. Some contracts set a minimum spend or a notice period.
What can go wrong
Cases signed this month often come from inquiries of earlier months. Count new inquiries and booked consults during the test, so you see the effect sooner.
Sources
- World Bank, Development Impact blog, summary of a research paper, Feb 24, 2014.