Lookalikes are dead. Interest stacks are back.
Why post-iOS Meta targeting forced us to re-learn manual interest layering for property developer audiences.
Lookalike audiences on Meta for UK property developers stopped being a reliable workhorse in late 2024. iOS-driven signal loss meant the seed audiences were noisier, the 1-3% lookalikes were drifting, and the cost-per-qualified-lead from a lookalike-led campaign was up roughly 60% on its 2022 baseline. By February 2025 we had stopped using them.
What replaced them
Manual interest stacks, layered three deep. 'Property investment' AND 'commercial real estate' AND ('Companies House' OR 'planning permission'). The kind of audience construction that was rendered obsolete in 2019 by lookalikes and quietly came back into its own through 2024.
These audiences are smaller, more expensive to deliver to, and the cost-per-impression is higher. They also convert dramatically better, because the intersection of three property-developer interests is a much purer audience than any lookalike seeded from a leaky pixel.
Why the math now works
In a world of clean pixel data and large training pools, a lookalike will beat a manual interest stack basically every time. In a world of degraded signal, the lookalike's edge collapses while the interest stack's targeting holds.
We were paying double the CPM and getting triple the qualified lead rate. Net cost-per-qualified-lead was meaningfully lower.
What this said about the channel
Meta in 2025 was rewarding the operators who were willing to do the older, more manual work. Advantage+ took care of the cold prospecting; manual interest stacks took care of the niche audiences that algorithmic targeting could no longer find cleanly.
The brokers winning the channel were treating it more like 2019 paid media than 2022 paid media.
Lookalikes are not gone forever - if the signal layer recovers they will come back into rotation. But for the specific problem of finding UK property developers on Meta in early 2025, manually-layered interest stacks were the better tool. Sometimes the older playbook is the right playbook.
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.