How Much Do Lawyers Spend on Marketing? Setting a Law Firm Marketing Budget

Ask what a law firm should spend on marketing and you will be told five to fifteen percent of revenue, with a straight face, by people who have never seen the study. I went looking for that study. It does not exist.

What does exist is enough evidence to build the number properly, from your own economics rather than from a percentage somebody made up. That is a longer conversation than a rule of thumb, and it produces a budget you can actually defend to your partners.

Before anything below is useful, one condition. If you cannot currently say what a signed client costs you, every number in this article is decoration. We have written separately about what a law firm lead should cost, and the tracking that makes that answerable comes before the budget, not after it.

How much do lawyers spend on marketing?

No, and the honesty is worth more than a fake number.

I checked every source that gets cited. Clio does not publish marketing spend as a share of revenue. The Law Society of England and Wales runs the best law firm financial benchmark in the UK, covering 121 firms and £1.2bn of fee income, and marketing is not broken out: it sits inside non-salary overheads, which fell to 28.4% of fee income in the 2025 financial year. Thomson Reuters' financial index tracks overhead growth but publishes no marketing line. The best Australian legal marketing benchmark I could find is a decade old, and it reports attitudes rather than spend.

The one law-firm-specific distribution I found with a stated method is a 2021 Justia reader survey, and it measures share of budget rather than revenue: 55% of firms put 0 to 5% of their budget into marketing, 23% put 6 to 10%, and almost none went above 30%. Self-selected sample, no size disclosed.

So when a page tells you five to fifteen percent of gross revenue, understand what you are reading. It is an agency FAQ answer that has been repeated until it acquired the texture of research.

The nearest defensible reference points are cross-industry. Gartner's 2025 survey of 402 marketing leaders, mostly at companies above $1bn of revenue, put marketing at 7.7% of revenue, unchanged on the year before, with paid media taking 30.6% of that. The Duke CMO Survey, fielded in early 2025 with 281 respondents, put marketing at 9.4% of revenue overall and 9.0% for business-to-business services.

Neither sample is a twelve lawyer firm. Use them as orientation, then build your own number.

What does a click and an enquiry actually cost?

More than in any other industry, and the variation inside legal is larger than the variation between industries.

The cleanest benchmark available is WordStream's 2026 study of 13,474 US search campaigns over the twelve months to March 2026. For attorneys and legal services it reports an average cost per click of $9.87, the highest of the 23 industries tracked, a conversion rate of 5.55%, and a cost per lead of $131.63. It is a vendor's own managed accounts, which skews small to mid budget advertisers, but the sample and method are disclosed, which is more than most.

That average hides the real spread. Using Google's own keyword data, median costs per click by case type run to $274 for motorcycle accidents, $321 for brain injury, $413 for truck accidents and $848 for offshore accidents, with individual keywords in some US cities touching $1,000.

Outside the US the picture is calmer. In the UK, the dearest legal terms sit around $30 a click for accident claim work, with conveyancing near $10 and criminal defence near $3.50. In Australia, TPD work runs near $19, workers compensation and general compensation near $14, and wills and estates near $2.

A warning about your own research here. Ahrefs, which is a good tool, returns figures like $0.45 a click for "personal injury lawyer" in the US. That is not real. If you are building a case on US legal click costs, use Google's own data or a platform benchmark, not a third-party estimate.

What does a signed client cost, and who do you believe?

This is where published benchmarks fall apart, and the disagreement is itself the finding.

For US personal injury paid search, one agency puts cost per lead at $700 to $1,500 and cost per signed case at $2,500 to $3,000, with lead to signed conversion of 10 to 15%. A second, publishing a study of 49 firms across 36 states, puts cost per lead at $442 for search, $378 for local services ads and $183 for organic. A third, using its own client accounts, says $500 to $700 is excellent for a car accident lead and up to $2,000 can still work, with conversion of 20 to 35% once optimised.

Three agencies, threefold disagreement, on the most basic number in the category. None of them is lying. They have different clients, different markets and different definitions of a lead.

In our own accounts, which are Australian and mostly personal injury and accounting, we see between 10% and 40% of enquiries turn out to be viable matters depending on the claim type and how hard the form screens, and when a lead is genuinely qualified the conversion to signed client sits nearer 70 to 80%. Those are our numbers, from our clients' feedback, and I would not ask you to plan on them either.

How do you work out what you can afford?

Backwards from the fee, in four steps. This is the arithmetic I use in every first conversation with a firm.

A worked example, using the US figures above so you can see it with real numbers. Average settlement in a widely cited reader survey was $52,900, and the average contingency was around a third, which is a gross fee near $17,500. A firm willing to spend 15% of that can pay about $2,600 to sign a client. At a 12.5% enquiry to signed rate, that is about $325 an enquiry, which buys you a fair amount of paid search in most case types but not truck accident work in an expensive city.

Run the same arithmetic in the UK and it tightens sharply, because the fee is capped rather than negotiated. Success fees on conditional fee agreements cannot exceed 100% of base costs and, for personal injury at first instance, are capped at 25% of general damages and past pecuniary loss. Damages-based agreements are capped at 25% of damages in personal injury. Since 2013 the success fee has not been recoverable from the losing party, so it comes out of the client's damages.

In Australia, ordinary personal injury work has no percentage-of-damages fee at all. Costs are time-based with an uplift capped at 25% of the legal costs otherwise payable, and Queensland adds a hard ceiling where professional fees cannot exceed 50% of the settlement after refunds and disbursements. So an Australian firm's affordable acquisition cost is a function of the cost of running the matter, not of the damages.

Same arithmetic, three different answers, which is why an imported US benchmark will mislead you.

What is the 70/20/10 rule, and does it apply to a law firm?

You will meet this rule in most budget articles, so it is worth knowing what it is and where it comes from.

The split is 70% of the budget on what is already proven to work, 20% on scaling things that show early promise, and 10% on experiments that will mostly fail. It is a general marketing heuristic borrowed from consumer brands, not a legal finding, and there is no study behind its application to law firms.

Used loosely it is sensible. It stops a firm putting everything into the channel that worked last year, and it reserves a small amount for the thing that becomes next year's main channel. In a firm spending $10,000 a month, the 10% is $1,000, which is enough to test a new claim type or a new format properly.

Where it misleads is at small budgets. If your whole budget is $3,000 a month, the experiment money is $300, which buys you noise rather than a test, and the 20% "scaling" bucket implies you already have something proven. Firms in their first year should put close to everything into one channel until it works.

Where should the money go once you have the number?

Split it between work that produces enquiries now and work that makes you the firm people think of later, and do not let the first one eat all of it.

The best known guidance here is Binet and Field's finding from the IPA databank that high performing campaigns cluster around 60% brand building and 40% sales activation. Two honest caveats. The sample is award entries, which means campaigns whose owners already believed they worked, and the authors present 60/40 as a benchmark and a starting point rather than a law. The industry has stripped both caveats away.

The adjacent idea that matters more for legal is the observation from the Ehrenberg-Bass Institute that at any moment only about 5% of business buyers are in market, with the other 95% not buying for months or years. For commercial and litigation practices that is the whole argument for content and profile work: most of the people who will brief you this year are not searching for you today.

For consumer practices the split leans harder to activation, because the trigger is an accident or a separation rather than a budget cycle, and the search happens within days of the event.

What should a small firm's first $2,000 a month buy?

This is the question almost nobody answers, and it is the one most Australian and UK firms are actually asking.

At that level you cannot run several channels, so the honest allocation is one channel plus the plumbing that makes it measurable. In practice that means a few hundred dollars of tracking and intake work in the first month, and the rest in paid search or paid social depending on whether your clients already know they need a lawyer.

What it will not buy is patience-free results. At $2,000 a month in a practice area where enquiries cost $200, you are buying about ten enquiries a month, which at a 20% sign rate is two matters. That is a real return in most practice areas and a very thin dataset for making decisions, which is why small budgets need longer before you judge them, not less.

What I would not do with the first $2,000 is spread it across search, social, a content writer and a directory listing. Four channels at $500 each produce four sets of numbers too small to read.

What happens if you cut the budget when things get tight?

It costs more than the saving, and this is one of the few areas of marketing with decent long-run modelling behind it.

The IPA's work on advertising in a downturn, drawn from a databank of 880 case studies since 1980, reports a rule of thumb that for every 10 points share of voice exceeds share of market, a brand can expect to gain about a point of market share a year, and to lose a point a year in the other direction.

Econometric modelling in the same body of work found that a budget cut to zero for one year, then restored, took five years for sales to recover, and a budget halved for one year took three. Separate profit modelling found short-term profit improves deceptively after a cut, because of the lag between spending and effect, then declines sharply in the third year.

None of that is legal-specific, and a firm with strong referral flow is more insulated than a packaged goods brand. But the shape of it is right, and it explains something I see regularly: firms that pause advertising in a slow quarter are usually still trying to climb back out of it two years later, and they attribute the recovery to the new campaign rather than to the hole the pause dug.

Where does the money actually get wasted?

Not where firms look for it. The largest documented waste in legal marketing is not in the media buy, it is in what happens after the enquiry arrives.

In a study of 1,333 US law firms, every one of which received a genuine enquiry through its own website form during business hours, 26% never responded at all. Median response time among those that did was 13 minutes, and a quarter responded within five.

Put that against a cost per lead anywhere between $130 and $1,500. A quarter of the budget is being spent to generate enquiries that nobody answers. There is no targeting change, bid strategy or landing page that recovers that money.

Clio's mystery shop of 500 firms found the same thing from the other side: 33% responded to an email, 40% answered the phone, and only 12% could estimate what a matter would cost.

The second waste is measurement. Firms that cannot connect spend to signed matters cannot tell a good channel from a bad one, so they optimise toward whatever is cheapest per lead, which is reliably the traffic least likely to become a client. We wrote about how that happens inside an account in why your Google Ads leads are junk.

What should your budget buy, month by month?

For a firm starting from nothing, in the order I would fund it.

The mistake I see most often is a firm spreading a modest budget across everything at once so that nothing produces enough data to judge. Small budgets are harder to manage than large ones, not easier, because everything sits inside the noise. We covered what that looks like over time in the first six months of law firm advertising.

How do you measure marketing ROI, and what KPIs should you hold it to?

By watching three things, monthly, and being willing to act on them.

Cost per signed client, against the affordable figure you calculated from the fee. If it is comfortably under, you are underspending and should push harder. If it is over, the problem is usually qualification or intake rather than the media.

The value of matters signed, not just the count. A month of cheap, small matters can look better than a month with two large ones and be worth a fraction as much. This is the number that stops a firm optimising itself into work it does not want.

The share of enquiries nobody reached. If that number is rising, no budget decision will help you until it comes down.

There is a line I use with clients that sounds flippant and is not. Nobody wants a fifteen dollar lead until they have a fifteen dollar lead. Then the same principal who asked for cheaper leads tells me he would happily pay a hundred and fifty for a better one. Both instincts are right at different moments, and the only thing that settles the argument is knowing what a signed client costs.

So what is the answer?

If you want a number to start from, take 7 to 9% of revenue as an orientation, understand that it comes from cross-industry research rather than from legal, and treat it as a hypothesis rather than a target.

Then do the real work, which is establishing what a signed client costs you and what a matter is worth, and adjusting until those two numbers sit in a sensible relationship. A firm that knows both can spend aggressively with a clear head. A firm that knows neither is guessing, whatever percentage it has chosen.

That shift is the whole point. It turns marketing from a cost you tolerate into a purchase you make deliberately, which is the position every firm I work with is trying to reach. If you want to see how we build that measurement, the growth system page covers the mechanics and PixelRush HQ is the tracking that sits underneath it.