How Do I Know If My Advertising Budget Is Being Spent Wisely?

Here is a situation that plays out in thousands of businesses every month. Google reports 40 conversions. Meta reports 35. Your accounting software shows 50 orders.

Nobody is lying. Both platforms counted every sale they touched, and most buyers touched both. The platforms are answering a question you did not ask: "did I contribute?" — when the question you need answered is "did the money work?"

This guide replaces platform-reported optimism with four numbers that reconcile to your bank account, and gives you an audit you can run this afternoon.

Key Takeaways

• Platform-reported ROAS almost always overstates performance because every platform claims credit for conversions it merely touched.
• Marketing efficiency ratio (MER) — total revenue divided by total ad spend — sidesteps attribution entirely. Benchmarks sit around 3x to 5x for most DTC verticals in 2026.
• Profitable ecommerce brands commonly run a blended MER between 2.5x and 4x; mature subscription brands push above 6x.
• The decisive test is not ROAS but contribution margin after ad spend. A 4x ROAS on a 20% margin product loses money.
• Incrementality testing — measuring what happens when a channel is switched off — becomes worth its cost above roughly £25,000–£50,000 in monthly spend.

What is the fastest way to know if my ad spend is working?

Divide your total revenue for last month by your total ad spend for last month. That single number — your marketing efficiency ratio, also called blended ROAS — tells you more than any platform dashboard, because it cannot be inflated by attribution.

MER = Total Revenue ÷ Total Ad Spend

If you did $80,000 in revenue on $20,000 of ad spend, your MER is 4.0. Every dollar of advertising is associated with four dollars of revenue.

Benchmarks for 2026 sit around 3x to 5x for most DTC verticals, with mature subscription brands pushing above 6x, and profitable ecommerce brands commonly running between 2.5x and 4x. Wherever you land, the trend line over six months tells you more than the absolute figure — MER falling while spend rises means you are scaling into inefficiency.

Why does platform-reported ROAS overstate performance?

Three structural reasons, none of which are the platforms behaving badly — they are answering their own question honestly.

Overlapping attribution windows

A customer sees your Instagram ad on Monday, searches your brand name on Thursday, clicks your Google ad and buys. Meta claims the sale within its view-through window. Google claims it as a click conversion. One sale, two claims. Sum the platform reports and you will exceed your actual revenue — frequently by 30–60%.

Branded search cannibalisation

A significant share of the conversions your Google brand campaign reports would have happened anyway. Someone typing your company name was already coming. That campaign often shows the highest ROAS in the account while being the least incremental spend in it.

Modelled conversions

Post-privacy, platforms estimate conversions they cannot directly observe. Modelling is reasonable in aggregate and unreliable at small volumes. At low spend, a meaningful share of reported conversions is a statistical inference, not an observed sale.

None of this makes platform data useless. Use it for relative decisions inside a platform, which ad set beats which, which creative won. Never use it to judge whether the business is making money. That is what MER is for.

What four numbers actually tell you the truth?

1. Marketing efficiency ratio (MER)

Total revenue ÷ total ad spend. Your headline efficiency number. Track it monthly and watch the trend.

2. Blended CAC

Total ad spend ÷ total new customers. Not per channel — total. This is the number to compare against your maximum allowable CAC, calculated from lifetime value at a 3:1 ratio. Method in the following article: [how much should I spend on advertising].

3. Contribution margin after ad spend

The number that actually decides whether you have a business. Take revenue, subtract cost of goods, subtract shipping and fulfilment, subtract payment processing, subtract returns, then subtract ad spend.

Worth being blunt about why this matters. A 4x ROAS sounds excellent. On a product with a 20% gross margin it is a loss:

Identical ROAS, opposite outcomes. Any agency reporting ROAS without knowing your margin is reporting a number that cannot tell you whether to keep spending.

4. CAC payback period

How many months of gross profit it takes to recover acquisition cost. Under three months, you can scale aggressively because the money recycles fast. Beyond twelve, growth is financed from working capital and scale becomes a cash-flow problem before it becomes a profit problem.

How do you run a 15-minute budget audit?

Do this monthly. It takes longer the first time because you have to find the numbers; after that it is quick.

Step 1: Reconcile platform claims to reality (3 minutes)

Add up conversions reported by every ad platform for last month. Compare to actual orders from your ecommerce or accounting system.

Read it: Platforms claiming 20–40% more than reality is normal. Over 60% means your attribution windows are too generous or view-through is doing too much work. Under 100% means tracking is broken and you are under-reporting genuine performance.

Step 2: Calculate MER and blended CAC (2 minutes)

Total revenue ÷ total spend. Total spend ÷ new customers.

Read it: Compare against last month and six months ago. Direction matters more than level.

Step 3: Check the learning threshold (3 minutes)

For each ad set, divide monthly conversions by 4.3 to get weekly conversions.

Read it: Ad sets under 50 conversions per week are not optimising properly — Meta's documented threshold. If most of your account sits below it, consolidation will improve results more than any creative change.

Step 4: Find your brand-search share (2 minutes)

In Google Ads, isolate spend on campaigns targeting your own brand name.

Read it: If branded search is a large share of spend and reported conversions, your account looks far healthier than it is. Most of that demand already existed. Consider a brand-campaign pause test to see what share you were buying unnecessarily.

Step 5: Audit creative age (2 minutes)

Sort ads by spend. Check the launch date of your top three.

Read it: Social creative fatigues in weeks. If your top spenders are months old with rising frequency and falling CTR, you are paying escalating prices for declining attention.

Step 6: Check audience overlap (3 minutes)

List your active audiences and look for ones that contain the same people — a broad interest audience and a lookalike built from the same seed, for example.

Read it: Overlapping audiences bid against each other in the same auction. You are competing with yourself and paying the premium.

Where does ad budget actually leak?

Seven patterns account for most waste in audited accounts.

1. Branded search you did not need to buy. Some defensive spend is justified if competitors bid on your name; buying 100% of your own brand traffic rarely is.
2. Ad sets permanently below the learning threshold. Paying for volatility, not performance.
3. Audience overlap. Self-inflicted auction inflation.
4. Retargeting windows that are too long. A 180-day window claims credit for people who forgot you existed. Test 7 and 14 days.
5. Placements you never chose. Automatic placements can push meaningful spend into low-quality inventory. Check the placement breakdown.
6. Geographic spill. Targeting set to "people in or interested in" a location delivers to people browsing from elsewhere. Set it to "people living in."
7. Optimising for the wrong event. Chasing cheap conversions on your lowest-margin product because that is literally what you asked the algorithm to find.

When should you use incrementality testing?

Incrementality testing answers the only question that truly matters: what would happen to revenue if this channel were switched off?

The cheapest version is a geo holdout. Pick two comparable regions, turn a channel off in one for two to four weeks, and compare revenue. If revenue holds steady, that spend was not incremental — you were paying for customers who were coming anyway.

Practically, most businesses run on blended MER for their first year or two, add contribution-based MER once they trust their margin data, and only invest in formal incrementality testing once monthly spend passes roughly $25,000–$50,000. Below that, the test costs more in lost revenue and analyst time than the insight is worth.

One exception worth making early: a branded search pause test. It is cheap, fast, and frequently the single most revealing experiment a small advertiser can run.

Frequently Asked Questions

What is a good MER for my business?

It depends on your gross margin. Divide 1 by your gross margin percentage to find break-even MER — at 50% margin, break-even is 2.0x, so anything above that contributes. General 2026 benchmarks sit at 3x to 5x for DTC, but your own break-even figure is the number that matters.

My agency reports 6x ROAS but I'm not making money. What's wrong?

Almost always one of three things: platform ROAS is double-counting across channels; the ROAS is concentrated in branded search and retargeting that would have converted anyway; or your margin cannot support the spend even at that ratio. Calculate MER and contribution margin after ad spend — those two numbers will identify which.

How often should I check these numbers?

MER and blended CAC monthly. Spend pacing weekly. Resist checking daily — daily data at small volumes is mostly noise, and reacting to it causes the learning-phase resets that damage performance.

Does MER work for lead generation, not just ecommerce?

Yes, with a substitution. Use closed revenue rather than pipeline value, and account for the lag between lead and close. If your sales cycle is 60 days, compare this month's revenue against ad spend from two months ago, or the ratio will look worse than it is during any growth period.

Should I stop using platform ROAS entirely?

No — use it for the right job. Platform data is reliable for comparing options within a platform: which ad set, which creative, which audience. It is unreliable for judging whether the channel deserves the budget. Optimise with platform data; decide with blended data.

The bottom line

Your platforms are not lying to you, but they are answering a different question than the one you need answered. Every one of them reports on its own contribution, and those contributions overlap.

Run the six-step audit this month. If your platforms are claiming more than about 40% above your real order count, if most ad sets sit below the learning threshold, or if your top-spending creative is three months old, you have found real money — and you found it without spending anything to look.

If the audit shows your budget is sound but results are still short, the constraint is probably allocation rather than execution

Sources

• Eightx, "What Is MER (Marketing Efficiency Ratio)? Formula + 2026 DTC Benchmarks" — MER benchmarks of 3x–5x for DTC verticals, 2026.
• Shopify, "Marketing Efficiency Ratio: How To Calculate + Improve MER (2026)" — MER formula and definition.
• Northbeam, "Marketing Efficiency Ratio (MER): Definition, Benchmarks, and How it Differs from ROAS" — blended MER range of 2.5x–4x for profitable ecommerce brands.
• Superscale, "MER vs ROAS in 2026: what each means and when to use it" — incrementality testing threshold of £25,000–£50,000 monthly spend.
• Meta Business Help Center — ad set delivery and learning-phase optimization guidance (~50 conversions per ad set per week; platform documentation, no single stable article URL).