Stop Bidding on Leads. Start Bidding on Case Value.

Here is a report that looks excellent and is quietly destroying a practice: cost per lead down 22% year over year, lead volume up 31%, and revenue flat.

It happens constantly, and the mechanism is always the same. The account is optimising for the number of conversions, so it finds the conversions that are easiest to produce. In a mixed legal practice, the easiest conversions to produce are almost never the matters that fund the firm.

One conversion, wildly different economics

Take an estates and probate practice. A simple will, a trust package, an uncontested probate and a contested probate might sit at $1,500, $4,500, $9,000 and $40,000 respectively.

By default, every one of those is exactly one conversion. The bidding system has no way to tell them apart, so it does the rational thing and pursues whichever is cheapest to acquire, which is the $1,500 matter, in volume.

The firm gets busier. The lawyers work harder. The revenue line does not move. And the report says everything improved.

What value-based bidding changes

Value-based bidding gives each conversion a number, so the system can optimise toward total value rather than total count. Google recommends using a conversion value calculator to establish the average value of your conversion actions before moving to a value-based strategy, and provides conversion value rules to adjust value in real time by dimension such as location (Google Ads Help).

For a law firm, “value” can mean several sensible things, and picking the right one matters:

• Expected value at enquiry: average matter value for that practice area, multiplied by the historic rate at which that enquiry type becomes a client. Useful early, and usable without CRM integration.
• Actual fee at engagement: the real number, uploaded once the matter is opened. More accurate, requires the feedback loop to be running.
• Expected lifetime value: for practices with genuine repeat or referral patterns, such as business and estates work.

The sequence, and why skipping ahead hurts

Step one, count-based. Target cost per action, on clean conversion actions where only meaningful events count as primary. Most accounts we inherit are not even here, because eleven different actions are all marked primary and bidding is chasing an average of everything.

Step two, static values. Assign an expected value per practice area. Crude, and enormously better than treating everything as equal. This step alone often shifts budget out of low-value keyword sets within a month.

Step three, values from the CRM. Actual matter values flow back with the outcome. Now the numbers are real.

Step four, value bidding. Maximise conversion value, with a return target, plus conversion value rules for the dimensions that genuinely differ, such as counties where the firm wins more work or matter types that consistently over-perform.

Firms want to jump to step four. Doing it on unreliable data is worse than staying at step one, because you have now taught the system something confidently wrong, and it will pursue that error efficiently.

The uncomfortable prerequisite

Value-based bidding requires the firm to be able to answer a question many cannot: what is a matter of each type actually worth to us, and how often does an enquiry of each type become one?

That is not a marketing question. It is a practice management question, and the conversation usually surfaces things the owner half-suspected: a practice area that generates enormous enquiry volume and almost no revenue, a referral source everybody assumed was valuable, a matter type where the close rate collapsed two years ago and nobody noticed.

We would rather have that conversation in month one than build an elegant bidding strategy on top of numbers nobody believes.

What it looks like when it works

The report stops being about leads. It becomes: what did we spend, what value did it produce, what did a signed matter cost us, and which practice areas are actually carrying the firm.

Budget moves toward the work you want more of. Sometimes that means fewer leads and a better year, which is a result you can only defend if you were measuring the right thing from the start.

Law Ops Forge builds the value model with the firm, from the firm’s own matter data, and revisits it quarterly as part of managed account work. Start with a conversation about what your matters are actually worth.

Sources

Every claim above comes from Google’s own documentation. These are the pages it came from, so you can check any of it yourself, or send it to whoever else needs convincing.

• Google Ads Help: value-based bidding and conversion value rules
• Google Ads Help: about Ad Rank
• Google Ads API: upload offline click conversions, where values are sent
• Google Ads Help: about offline conversion imports