Services

Book a call
2 new clients / monthmaik@starkdigital.ie

Legal · Dublin 8, Ireland · November 2025 to September 2026

Paid search the firm could not measure, and what changed when we took the account

How Anthony Joyce & Co. Solicitors went from switching Google Ads off entirely to 596 tracked enquiries in ten months, with cost per enquiry down 80 per cent.

596

Tracked enquiries

−80%

Cost per enquiry

4.8×

Budget the firm added

€2.37

Cost per click

Each of these is set out with its measurement basis further down. Nothing on this page is quoted without one.

The pages behind it

Screenshot of immigration.anthonyjoyce.ie

immigration.anthonyjoyce.ie

Citizenship, work permits and IRP renewals, built around the phone number rather than a form. Takes the majority of the account’s traffic.

Screenshot of notary.anthonyjoyce.ie

notary.anthonyjoyce.ie

AHPRA document certification for nurses and doctors registering to practise in Australia. A tiny market, with very few firms serving it.

A Dublin law firm had been running Google Ads for years without a reliable way to tell what the campaigns were producing. In July 2025, with no evidence either way, they switched the ads off.

We took the account on 25 November 2025. Ten months later it had produced 596 tracked enquiries, the cost of each one had fallen from over €100 to under €22, and the firm had at one point increased its budget almost fivefold.

This is what we did, in the order we did it, including the month it went badly.

Four numbers, and what they actually measure

596

Tracked enquiries

Calls and form submissions matched to a paid click by WhatConverts, 25 November 2025 to 19 September 2026. An enquiry is not a client. Whether one becomes a client is the firm's decision, not ours.

−80%

Cost per enquiry

Google Ads spend divided by tracked enquiries, first month against most recent month. €107 in November 2025, €22 in September 2026.

4.8×

Budget the firm added

Monthly Google Ads spend, lowest month of the engagement (March 2026) against highest (July 2026). Every increase was requested by the firm, not a change we made on their behalf.

€2.37

Cost per click

Average cost per click under our management, against €4.48 across the account's history before it. Same account, same market, same firm.

Every figure on this page comes from one of three places: the firm's own monthly marketing reports from January 2024 to August 2025, the Google Ads account, or WhatConverts, the call and form tracking we installed. Where a figure is modelled rather than measured, it says so.

What we found when we arrived

The firm had been running Google Ads since at least 2022, and kept detailed monthly marketing reports, which it gave us access to. What those reports could not do was connect the ad spend to the enquiries it produced. The previous campaigns could not be measured, so nobody could say with confidence whether they were working.

In July 2025 the firm switched the campaigns off.

The reports for the following weeks record what happened next: new enquiries across the whole firm dropped sharply, and the reports attribute the drop to turning the ads off. Their recommendation was to restart paid advertising.

So this was the situation in November 2025: a firm that knew paid search mattered, because switching it off had hurt, and that had no way yet to see what it was getting for the money.

The problem underneath the problem

Read the reports closely and one thing comes up again and again. The firm could not see where its enquiries were coming from.

In February 2025, litigation clients appear in the firm's lead log but not in the tracking software, so nobody can tell whether they came from ads or from organic search. In April and May, the reports say paid campaigns are generating traffic but nobody can confirm whether any of it turns into client files. In July 2025, a lapse in conversion tracking leaves nine new clients with no known source at all.

For eighteen months the firm was paying for something it could not measure, and so nobody could fix it. That is where we started.

We also looked at the change history on the old campaigns. Each of the five main campaigns carries around 700 recorded status changes, almost all of them switching the campaign on or off. That pattern is consistent with automated scheduling rather than manual optimisation, and a campaign switched on and off that often never accumulates enough continuous data for Google's bidding to learn from.

Step one: fix the measurement, not the campaigns

The first three weeks were spent on tracking, not on advertising.

We connected WhatConverts to the firm's phone system and website forms, so that every call and every form submission could be matched to the search that produced it. We fixed a mismatch between Google's iOS click identifiers and the tracking software that had been silently dropping a share of conversions. We set up the phone routing so that ad-driven calls could be told apart from everything else.

None of this produces enquiries. It is unglamorous, and in the first month the firm could see nothing for it. But it meant that from December onwards every euro spent could be traced to what it produced, and every decision after that was made on real numbers rather than on how a month felt.

The first month, November 2025, cost €643 and produced six tracked enquiries at €107 each. That figure is in the table below, because it is the honest starting point.

Step two: one campaign per practice area

The old account ran broad campaigns with names like "Solicitors in Dublin" and "No Win No Fee". A firm that handles immigration, personal injury, medical negligence and notarial work does not have one type of client. It has four, searching for different things, worth different amounts, converting at different rates.

We rebuilt the account with a separate campaign for each practice area, each with its own budget, its own keywords, its own ads and its own landing page. This is more work to set up and more work to run, and it is the only way to find out which parts of a law firm's advertising actually pay.

Ten months on, the answer is clear. Immigration produces enquiries at roughly a third of the cost of medical negligence. That is not a criticism of either campaign. It is information the firm did not have before, and it now decides where the budget goes.

Step three: pages built for the search, not the firm's homepage

The old campaigns sent every click to the firm's main website. Somebody who has just searched for "citizenship application solicitor" and lands on a general law firm homepage has to find their own way to the right page, and most do not.

We built dedicated landing pages for the immigration and notarial campaigns, each written for the specific search that brings the visitor there, with one clear way to get in touch and nothing else. The pages sit on the firm's own domain, so the trust of the firm's name carries through.

The immigration page has been live for most of the engagement and takes the majority of the account's traffic. Its click-through rate of 8.6 per cent across the full period, rising above 15 per cent in September 2026, is several times the account's previous average.

Step four: the campaign nobody else would have built

Irish nurses and doctors who want to register to practise in Australia need certain documents certified in a specific way that the Australian regulator, AHPRA, will accept. A notary who understands exactly what AHPRA requires is hard to find. The people looking for one are on a deadline, have few alternatives, and search with unusual precision.

This is a tiny market.

46

Clicks in ten months

Google Ads, AHPRA notarial campaign, full period. A tiny market, deliberately.

€10.90

Cost per click

Campaign average, full period. High, and worth it: each enquiry is a specific, time-sensitive piece of work with almost no competition.

Forty-six clicks in ten months is not a business on its own. It does not need to be: each enquiry is a specific, time-sensitive piece of work, and the people making them have very few places to go.

No keyword tool suggests AHPRA notarial certification. You find it by asking the firm what work it does that nobody else advertises.

What happened

Spend against cost per enquiry

Relative spendCost / enquiry
€0€55€110€107€86€22Nov2025DecJan2026FebMarAprMayJunJulAugSep

The bars go up because the firm asked for more budget. The line comes down anyway. April is the exception, and July is the month they asked us to stop.

Google Ads account, November 2025 to 19 September 2026. Bars show relative spend, the line reads against the right axis. Every cost per enquiry is in the table below.

Clicks, tracked enquiries and cost per enquiry by month, November 2025 to September 2026.
MonthClicksEnquiriesCost / enquiry
Nov 20251906€107
Dec 20251,10167€42
Jan 202692265€47
Feb 20261,73545€49
Mar 20261,20330€50
Apr 202656224€86
May 202688242€63
Jun 20262,07641€94
Jul 20262,347130€54
Aug 20261,16370€51
Sep 202690277€22
Ten months13,083596€52
  • Nov 2025 — Tracking build. Five weeks of spend, three of them before the new campaigns existed.
  • Apr 2026 — The bad month. We were rebalancing the personal injury campaign and got it wrong for about three weeks.
  • Jul 2026 — Peak. 130 enquiries. From August the firm chose to scale the budget back.
  • Sep 2026 — To the 19th. Half the budget of July, and the cheapest enquiries of the engagement.

Google Ads account, 25 November 2025 to 19 September 2026. A tracked enquiry is a call or form submission matched to a paid click by WhatConverts. September is a partial month. The monthly column sums to 597 against an account total of 596, because conversions attributed across more than one click leave half figures in the underlying data.

Look at the bars first. From March to July the firm more than quadrupled its monthly budget. Nobody quadruples an advertising budget on trust. They do it because the previous increase paid for itself and there is room for more. That is the clearest evidence on this page that the account was working, and it is the firm's decision rather than our claim.

Then read the last column of the table. Cost per enquiry started at €107 and finished at €22. The improvement is not a straight line, and April is a genuinely poor month: fewer clicks, fewer enquiries, and the worst cost per enquiry since the first weeks. We were adjusting the personal injury campaign at the time and got the balance wrong for about three weeks. It is in the table because a table without a bad month is not a real table.

The month we turned it down

July was the peak: the highest spend, and with 130 tracked enquiries the strongest month by a distance. After it, the firm chose to scale the budget back.

So from August we brought the immigration budget down in steps, and the account's spend fell by half. Cost per enquiry still fell to its lowest level yet, €22 in September, because the account by then had ten months of continuous data and Google's bidding had learned which searches convert.

The measure of a campaign is not whether it can produce more enquiries. It is whether it produces the number the firm can actually turn into work, at a cost that makes sense, month after month.

The first ninety days, for comparison

The figures quoted on the front of this site are from the first 90 days of this engagement, because that is the period a prospective client is deciding about.

+112%

Tracked enquiries

First 90 days after launch against the 90 days of the firm’s previous paid search period. Form submissions and phone calls, Google Ads.

−38%

Cost per enquiry

Same two 90-day periods, Google Ads spend divided by tracked enquiries.

3.4×

Return on ad spend

Modelled, not measured: signed cases attributed to Google Ads, valued at the firm’s own average case value, divided by spend over the same period. Specific to this account, and not quoted as a general expectation.

What we would say to another firm

Most of what worked here is not clever. It is the order things were done in.

Measurement first, because without it every other decision is a guess. One campaign per practice area, because a law firm is several businesses under one name. A page built for each search, because the firm's homepage is written for people who already know the firm. And a willingness to turn the budget down when the firm says it has enough, because the campaign exists to serve the firm and not the other way round.

Paid search worked for this firm once it could be measured. Everything else followed from that.

What the firm said

Working with Maik was incredibly smooth. He understood what we needed right away and we have already noticed more client enquiries coming in.
Anthony Joyce, Owner at Anthony Joyce & Co. Solicitors

Anthony Joyce

Owner · Anthony Joyce & Co. Solicitors

Anthony Joyce & Co. Solicitors, Dublin 8, Ireland. Figures last updated 19 September 2026. This page is refreshed with new monthly figures each quarter.

Next step

Thirty minutes, and an honest answer.

Tell me what you sell and what a customer is worth to you. I will tell you whether search is worth your money, roughly what it would take, and if the answer is no, what I would do instead.