Meta in 2026: the account structure that actually compounds
Advantage+ vs. manual, signal-loss budgets, and how we structure accounts that have to survive a KYC funnel.
Meta account structure debates have a way of turning into religious arguments — Advantage+ true believers on one side, manual-campaign holdouts on the other, each pointing at a case study that proves their side. Having now run this across 40-plus accounts spanning finance, D2C, quick-commerce and edtech, the honest answer is that the debate is mostly asking the wrong question. The structure that compounds isn't purely Advantage+ or purely manual — it's a specific division of labour between the two, and getting that division wrong costs more than picking either extreme.
What Advantage+ is actually good at
Advantage+ Shopping campaigns are genuinely strong at broad, prospecting-stage delivery, where Meta's own signal — increasingly rich, even as third-party signal degrades — can find incremental audiences a manually built lookalike would miss. Where it struggles is anything requiring deliberate creative-to-audience matching, or budget control at a granularity the automated system doesn't expose. Treating it as a replacement for the whole account structure, rather than a specific tool for a specific job, is where most of the disappointing Advantage+ results we've seen come from.
Where manual campaigns still earn their keep
Manual campaign structures remain the right call for retargeting, for lookalike audiences built off high-value customer segments the automated system has no way to identify as high-value, and for any creative test where you need to isolate which specific variable moved the needle — Advantage+'s automation actively works against clean testing, because it's optimising multiple variables simultaneously by design. The accounts that perform best in our portfolio run Advantage+ for top-of-funnel prospecting and manual structures for everything requiring precision, rather than picking one philosophy and forcing every campaign type through it.
Budgeting for signal loss instead of hoping it doesn't happen
The other structural decision that matters more in 2026 than it did a few years ago is building signal-loss recovery into the account from day one rather than treating it as a separate project. That means Conversions API deployment paired with the pixel from the start, not bolted on after performance degrades — and it means budgeting explicitly for the modelled-conversion gap Meta's own reporting will show once ITP and consent gaps are accounted for, rather than being surprised by it during a quarterly review.
Accounts that treat signal recovery as core infrastructure, standard from account setup rather than a remediation project after something breaks, consistently show more stable performance through the platform changes Meta ships several times a year. The ones treating it as optional are the ones we see scrambling every time Meta announces another attribution-window change.
The bid strategy that ties it together
Underneath both the Advantage+ and manual layers, we run portfolio-level budget optimisation that shifts spend between the two based on incremental return measured against actual revenue — not in-platform ROAS, which both structures will report favourably for very different reasons. That portfolio view is what actually decides, week to week, how the prospecting and retargeting budgets split, rather than a fixed ratio set once and left alone. Structure is only half the answer; the other half is a measurement layer honest enough to tell you when the structure needs to shift, which is the part most accounts we inherit are missing entirely.
