Paid Advertising for Professional Services & B2B — The Practitioner’s Playbook.

SectionHonest reframe

Most paid agencies sell accountants, lawyers, MSPs, financial advisers, HR consultancies and surveyors the same playbook they sell a plumber. Broad-match Performance Max, a "lead form" on a one-page landing site, a £20-a-month pixel and a monthly impressions report. Three months in, the partner who signed it off looks at the bank account, sees no pipeline, and quietly cancels.

B2B professional services paid is structurally different. The buying committee is three to five people — typically a finance lead, an operations lead, a technical or compliance specialist, and a sponsor partner or director. The cycle is three to nine months from first ad impression to first invoice raised. The conversion event is not a "lead", it is a marketing-qualified lead that is then sat with by a partner before it becomes a sales-qualified lead that is then scoped before it becomes a fee. LinkedIn ABM (account-based marketing) is the precision instrument; Google Search is the trigger-event capture; YouTube is the trust accelerator. Performance Max is, charitably, a tertiary tool for this category — and yet it is what generic agencies default to because it is what their automation knows how to run.

The other thing the generic shop misses: gated thought-leadership content is the unit economics of the category. A £40 cost-per-download list-build, sequenced into a six-month nurture, is the difference between paying £80 for an MQL and £350 for an SQL. Without it, every campaign is naked direct-response and the maths breaks at partner-level fee scale. The third miss is regulator-aligned creative — ICAEW, ACCA, CIMA, SRA, FCA, ICO and RICS each carry a credibility signal a buyer reads in under a second; if your accreditations and regulator memberships are not on the ad and the landing page, you are bidding from behind.

This playbook fixes the structure. Read it, run it yourself, or have us ship it on retainer. Same canon either way.

SectionEight-point audit

Score your own paid account red / amber / green this week.

1. LinkedIn ABM company-list targeting — Are you uploading a named target-account list (300–3,000 companies, by sector, size, region, regulator) and running campaigns specifically against named decision-makers in those companies? Or are you running broad LinkedIn "interest" targeting? The former produces SQLs; the latter produces newsletter sign-ups. Most professional-services accounts have NO uploaded company list.
2. MQL → SQL definition with a named scoring model — Is "MQL" defined (job title threshold, company-size threshold, content-engagement threshold) and is "SQL" defined (booked-call, scoping engagement, BANT met)? Is there a written scoring model? If MQL = "anyone who downloaded a PDF", the algorithm is bidding on the cheapest bottom-of-list noise.
3. Gated thought-leadership content as primary lead-magnet — Is there a named, current, sector-specific gated asset (R&D claims guide, regulatory-change briefing, year-end planning paper, post-merger integration checklist)? Is it refreshed quarterly? Generic "free consultation" CTAs convert at a fraction of the rate of a specific, dated, sector-specific paper.
4. Per-sector creative (manufacturing / professional / retail / public sector) — Are ad sets segmented by the client's sector, with creative, case-references and CTA tuned per vertical? Or is the creative the same generic "trusted advisers" stock photography? Sector-specific creative routinely outperforms generic by 40–80% on click-through and 2–3× on MQL conversion.
5. LinkedIn Lead Gen Forms + Conversation Ads — Are LinkedIn Lead Gen Forms running with pre-filled fields and immediate value-handoff (download in-platform), and are Conversation Ads running for higher-intent re-targeting against known target accounts? These are LinkedIn's most efficient formats for B2B and most accounts are not running them.
6. Offline conversion tracking enquiry → consult → engagement → fee billed — Is enquiry → booked-consult → scoped-engagement → fee-billed flowing back into Google + LinkedIn as offline conversions with revenue values? Without this the algorithm bids toward enquiry volume, which is junk. With it, the algorithm starts bidding on revenue-shaped customers within 60 days.
7. Brand-spend ratio under 15% — Sum of cost on brand-bid keywords (firm name, partner names, near-brand misspellings) ÷ total cost. Anything over 15% is paying Google to send people who already know your name. For mid-size firms, this is one of the largest leaks we find on first audit.
8. YouTube + Google Search for trigger-event keywords — Are you bidding on trigger-event keywords (acquisition advisers, R&D claim windows, year-end tax planning, ICO breach notification, post-merger integration, GDPR audit), and is YouTube running for upper-funnel awareness against decision-maker audiences? Trigger-event Search is the highest-intent volume in the category; YouTube is the cheapest decision-maker reach.

Three or more reds — fix the foundation before scaling spend.

SectionSix deliverables

LinkedIn ABM company-list targeting. Named target-account list of 300–3,000 companies built to sector / size / region / regulator criteria, uploaded to LinkedIn Matched Audiences, layered with decision-maker job-title targeting (CFO, FD, GC, CIO, COO, head of HR, head of operations, head of risk, partner). Separate ad-sets per sector. Creative referencing the buyer's actual industry, with regulator badges (ICAEW, ACCA, CIMA, SRA, FCA, ICO, RICS) where they apply. Reach is intentionally narrow and intentionally expensive — this is precision, not volume. The output is not "leads", it is named buyers at named accounts engaging with named content. Time to first signal: 21 days.

MQL → SQL scoring + creative. Written scoring model (job title × company size × engagement depth × sector fit × regulator-relevance) producing MQL → SQL → SAL gates with explicit threshold scores. Creative tuned to each gate: top-of-funnel awareness, mid-funnel gated content, bottom-funnel "book a 30-minute call with a named partner". Offline conversions wired back to platforms at each gate so the algorithm bids on revenue-shaped customers, not download-shaped customers. Partner-time on unqualified consults drops materially. Junk-MQL rate typically drops 40–60% within 90 days. Time to first signal: 30 days.

Per-sector campaign architecture. Separate campaigns per client-sector (manufacturing, professional services, retail, public sector, healthcare, real estate, financial services, third sector). Each with sector-specific keywords, sector-specific gated content, sector-specific case references, sector-specific landing pages and sector-specific testimonial pulls. Ad relevance scores improve, CPC drops 15–25%, MQL → SQL conversion lifts materially. The principle is simple: a manufacturing FD reading an ad that opens "as a manufacturing FD, you are dealing with…" converts at multiples of the rate of an ad that opens "trusted advisers since…".

LinkedIn Lead Gen Forms + Conversation Ads. Lead Gen Forms with pre-filled fields running across awareness and mid-funnel for frictionless gated-content downloads; Conversation Ads running against known target accounts and re-targeting pools, simulating a one-to-one DM with the partner or specialist who would actually take the call. Conversation Ads carry the highest reply-rate of any LinkedIn format and are the single fastest path from impression to booked-call in this category — but they only work if the named sender is a real, credible practitioner with a profile a buyer can audit, not a marketing-team alias. Time to first signal: 14 days.

Server-side + offline conversions enquiry → fee billed. GA4 + sGTM + practice-management / CRM matching keys + offline conversion uploads pushing enquiry → consult-booked → consult-attended → engagement-scoped → engagement-signed → fee-billed values back to Google, LinkedIn and Meta. The algorithm bids on real fee-paying clients, not form-fillers. Revenue values flow back to the bidding model so a £25,000 corporate restructuring engagement is weighted against a £400 self-assessment download accordingly. This is the single highest-leverage technical change in the category and it is missing on most professional-services accounts.

Trigger-event Google Search overlay. Pre-built Google Search campaigns aligned to trigger-event keywords — acquisitions, disposals, R&D claim windows, year-end tax planning, ICO breach response, FCA application, RICS valuation, post-merger integration, partner-rebrand, regulatory-investigation response. Concentrated burst budgets at known seasonal peaks (year-end, R&D claim deadlines, FY filing windows, regulatory-cycle deadlines for niche advisory specialisms). Reporting overlays trigger-event calendar on enquiry-volume charts so partners can see which seasonal pushes are paying back. Time to first signal: 7 days.

SectionWhat to do this week

Three actions, ranked by leverage. Same first three steps we ship in week one of a Foundation retainer.

1. Pull last-90-day search-term and LinkedIn audience reports. Find the brand-spend ratio and the top 50 wasted queries. Owner: marketing manager or office manager. Time: 60 minutes. The waste is in plain sight. If brand-spend is over 15%, that is your second-biggest leak. If LinkedIn is running on broad interest targeting with no uploaded company list, that is your biggest leak.
2. Write the MQL and SQL definitions on one page. Owner: managing partner or head of practice + marketing manager. Time: 90 minutes. Job-title threshold, company-size threshold, sector fit, engagement depth, scoring model. Until this is on paper, every paid campaign in the account is bidding blind.
3. Decide DIY, DWY or DFY for the next 90 days. Owner: managing partner. See the three ways.

SectionFive questions

What's the right LinkedIn vs Google split for our practice? Indicative split for most professional-services practices: 50–65% LinkedIn (ABM, Lead Gen Forms, Conversation Ads, Sponsored Content), 25–35% Google Search (trigger-event keywords, brand defence under 15%, sector terms), 10–15% YouTube (decision-maker awareness, retargeting). If your buyer is genuinely high-intent search-driven (e.g. urgent ICO breach response, urgent R&D claim before deadline), tilt Google heavier. If your buyer is awareness-driven (e.g. partner-led advisory, post-merger integration), tilt LinkedIn heavier.

What's a realistic cost-per-lead in our category? Indicative ranges, gated-content MQL → booked-consult SQL: accountants £40–£150 MQL, £200–£600 SQL; lawyers £80–£300 MQL, £400–£1,200 SQL; MSPs £60–£200 MQL, £350–£900 SQL; financial advisers £80–£250 MQL, £400–£1,000 SQL; HR consultancies £50–£180 MQL, £300–£800 SQL; surveyors £60–£200 MQL, £350–£900 SQL. CPL is leading; cost-per-acquired-fee-paying-client is the truth — typically 4–8× the SQL number across the category.

How much does an MQL → SQL scoring model actually move the numbers? Materially. Without scoring, the platform bids toward the cheapest enquiry — which is almost always a junior buyer at a sub-target company downloading a paper out of curiosity. With a written scoring model wired into offline conversions, the platform bids toward decision-makers at target-fit companies. Junk-MQL rate typically drops 40–60% within 90 days; partner time spent on unqualified consults drops proportionally.

Does the gated-content list-build maths actually work? Yes, if you do it properly. Indicative maths: £40 cost-per-download → 2,500 downloads/year on £100k spend → 6-month nurture sequence → 3–6% convert to booked-consult → 30–50% of consults convert to scoping engagement → 60–80% of scoping engagements convert to signed fee. Average fee value × conversion rate × volume = the ROAS. On a £15,000 average first-year fee for a mid-market accounting practice, that is a 5–8× return on paid spend at the back end of the funnel — but only if the front of the funnel is built right. The maths only works if (a) the gated content is actually good, (b) the nurture sequence is written by someone who knows the sector, and (c) MQL → SQL is wired into offline conversions so the algorithm learns. Skip any of those three and the maths breaks.

Can we run this ourselves with the playbook + £750 audit? Yes — most mid-size practices can run the campaign architecture themselves with a competent in-house marketing manager and a partner willing to give 90 minutes a fortnight to creative review. The £750 audit gives you a written red/amber/green of your current account, a named-owner / dated next-steps list, and the MQL → SQL scoring template. Credit toward first cycle if you sign for DWY/DFY within 30 days.

SectionWhere to go from here

If you want this shipped end-to-end on a productised retainer, book a 30-minute discovery call.

If you'd rather have a senior practitioner reviewing your team's paid spend each week, the coaching plans start at £750/month. If you have a hard deadline (a partner rebrand to support, a post-merger ad-account unification, an R&D claim window, a year-end push), the two-week embedded sprint lands a senior practitioner in your account for ten working days at £3,000 fixed.

Or run it yourself. Eight-point audit + one deliverable a month + twice-quarterly office hours.