The Meta ads playbook for bridging brokers, 2025 edition
Creative, audience and offer structures that were still printing for UK bridging firms heading into spring.
Meta in Q1 2025 was a meaningfully harder channel than it was a year earlier. Advantage+ had eaten most of the manual campaign structure, iOS attribution was still degraded, and financial services creative approvals had tightened again in February. Despite all of that, a tightly-run bridging account was still the cheapest source of qualified developer leads we had. This is the playbook we were running.
Creative: founder-led, not stock
Every winning creative in the cohort this quarter was either founder-on-camera or testimonial-from-broker. UGC-style direct address with subtitles, 15-22 seconds, opening with a specific developer pain point - 'your refurb completes in eight weeks but your lender wants twelve' - not a brand statement.
Static was almost entirely dead for cold acquisition at sensible CPMs. The two static formats still working were spec-sheet style 'rate / LTV / term' overlays and screenshot-style testimonials. Everything else underperformed video by 3x or worse.
Audience: stop fighting Advantage+
We spent most of 2024 trying to outperform Advantage+ with manual audience structures. By Q1 2025 we had stopped. The model wins on a sensible budget with a clean creative pool and properly set up conversion events.
The exception was a single retargeting layer for warm site traffic, which still earned its keep at roughly 4-5x ROAS through the quarter. Cold prospecting belonged to Advantage+. Warm belonged to manual.
Offer: lead the calendar, not the rate
Bridging accounts that led with a rate ('from 0.7% per month') were getting cheaper leads of much worse quality - rate-shopping developers with no deal. Accounts that led with a calendar booking ('15 minutes with a senior broker this week') paid more per lead but converted to drawn-down deals at roughly 3x the rate.
The right metric for the channel was always cost per drawn-down deal, not cost per lead. Anyone optimising the latter was burning money.
Conversion setup
Server-side via Conversions API, dedup with a deterministic event_id, calendar booking as the primary optimisation event, and a custom event firing on broker-confirmed qualified status downstream. Anything less than that and you were optimising on noise.
The accounts still working in Q1 2025 all had this infrastructure. The accounts struggling mostly didn't.
This playbook had maybe one more quarter of life in it. By the time we wrote the memo in April we were already running the AI outbound pilot that became our main business a month later. But for a brokerage with the right creative team and a real conversion stack, Meta was still a working channel in spring 2025. Just barely.
Capital Edge publishes one note a month on UK bridging finance, paid acquisition, and AI-led outbound. Written for brokers, by the team running the playbook.