How to run paid search in expensive categories without funding other people’s research.
Written for Firms and companies spending on search where clicks are costly and lead quality is inconsistent.
The bottleneck
In legal and high-value B2B, a single click can cost more than a month of hosting. Most accounts spend a meaningful share of that budget on people who are researching, comparing or writing a dissertation, because nothing separates those queries from hiring queries.
The second failure is measurement. Spend is reported against leads, leads are counted as form fills, and nobody connects either to signed work. That makes it impossible to tell an expensive campaign that works from a cheap one that does not.
Why it happens, five levels down
- Cost per lead is high and quality is inconsistent.Why? Research traffic is being paid for at hiring-traffic prices.
- Research and hiring queries share campaigns.Why? Campaigns were structured by service, not by intent.
- Negatives were never built out properly.Why? Nobody reviewed the search terms report after launch.
- Optimisation stopped after setup.Why? The account was treated as a build rather than an operation.
- Paid search has no operating rhythm.Root cause. It is a weekly discipline being managed as a quarterly project.
The diagnostic
Run this before planning any work. It tells you whether this is your constraint or whether something else is.
- Pull the search terms report for the last ninety days. Mark every term that is research rather than hiring.
- Check whether campaigns are split by intent, or only by service and location.
- Confirm calls are tracked and attributed, not just form fills.
- Check whether any CRM outcome flows back into the ad platform.
- Look at whether each ad group has a matching landing page or points at a generic one.
- Check geographic settings against where you can actually take work.
The build sequence
| Stage | Work | Timing |
|---|---|---|
| Separate | Split campaigns by intent before anything else. Research and hiring never share a budget. | Week 1 |
| Exclude | Build the negative list from the search terms report, then review it weekly. | Weeks 1–2, then ongoing |
| Match | Give each ad group a landing page that answers its specific query. | Weeks 2–3 |
| Measure | Track calls and forms, and connect them to CRM stages. | Week 3 |
| Scale | Increase spend only on themes with proven downstream outcomes. | Ongoing |
KPI watchlist
| Metric | Source | Read it as |
|---|---|---|
| Qualified leads | CRM stage, not form fills | The real numerator |
| Cost per qualified lead | Spend against CRM | The only honest efficiency measure |
| Consultation rate by campaign | CRM | Which themes deserve more budget |
| Signed matters or closed deals | CRM | What the spend is actually for |
Record a baseline and a date range for each before any work begins. Reconstructing a baseline afterwards is usually impossible.
Four ways this goes wrong
- Optimising to cost per click. It rewards cheap, useless traffic.
- Sending all traffic to the homepage.
- Running broad match without a maintained negative list.
- Reporting form fills to leadership as leads.
Get a search terms review
Most accounts have a list of expensive terms that have never produced work.
