The 20–40% number, explained honestly

You have seen the claim on this site’s homepage: 20–40% of Meta ad spend is wasted on average. Every tool in my category quotes some version of it. A number that convenient deserves scrutiny — so let me give my own marketing the audit treatment.

Where the range comes from

It is not one study; it is the overlap of several kinds of evidence. Industry audits of e-commerce ad accounts repeatedly land in this band when they total up four leak families: spend on audiences past saturation, spend on creative past its fatigue point, spend misallocated across campaigns relative to results, and spend optimized against broken or partial tracking signal. Each family alone is single digits to low teens; stacked, they reach 20–40% in a typical unmanaged or lightly-managed account. The width of the range is the honesty — accounts vary enormously.

When the number is wrong

• Tightly-run accounts: a disciplined operator who already caps frequency, refreshes creative on schedule and reconciles tracking can sit well under 20%. The number describes the typical account, not the best one.
• Tiny accounts: under a few hundred dollars a month, the variance of small numbers swamps everything. A range like this is meaningless there.
• Accounts with broken measurement: if tracking is badly wrong, you cannot even compute waste reliably — the first fix is signal, not spend.

How I compute YOUR number

When I audit an account I do not apply the benchmark — I replace it. Every leak I flag is sized from the account’s own history: what this ad set spent while frequency sat above the saturation line, what this creative spent after its click-through decayed past the fatigue threshold, what this campaign drew while converting below the account’s own median. Sum the line items and you get a number with receipts. Sometimes it is 9%. Sometimes it is worse than 40%. The benchmark’s only job is to tell you the question is worth asking.

The test worth running

Distrust any tool — including me — that quotes the range without showing line items. The 20–40% figure earns its place as a prior, not a promise. If a vendor cannot decompose your waste into named ad sets with dates and amounts, they are quoting the industry, not your account. The free audit exists precisely so you never have to take the headline number on faith.