DataOrganizer

Before the Q4 peak, agencies propose higher budgets, cost-per-click rises and the shop’s average MER still looks healthy. The average, however, hides the effect of the latest increase. In the example photography shop, the last €2,500 of spend brought in only €7,500 of revenue and reduced profit after advertising. This guide answers the question of when to increase the ad budget in an online shop. It includes the formula for incremental MER, a threshold derived from margin, a calculation at four spend levels and three decision thresholds. After reading it, you will be able to work out from your own data whether the next €250 still pays off, and use the same calculation to check your agency’s proposal.

Instead of the average, one question matters: how much shop revenue did the latest budget tranche bring in, and did the margin on that revenue cover its cost? With a 20% contribution margin, every additional €250 must generate at least €1,250 of revenue (1 / 0.20 = 5.0). Below that line, spend grows faster than profit, even when the average MER of the whole budget is still above the threshold.

The scene before the season

Average MER looks healthy, but the last tranche loses money

June in a photography shop, time to plan the season. The shop sells cameras, lenses, tripods and memory cards. In the latest 28-day period it spent €15,000 on advertising and generated €93,750 of revenue. The agency proposes raising the Q4 budget to €17,500 a month, because “the results are good”.

At first glance, the argument holds up. MER (shop revenue divided by total ad spend across all platforms) is €93,750 / €15,000 = 6.25. With a 20% contribution margin, the break-even threshold is 1 / 0.20 = 5.0, so the headroom looks comfortable.

The problem only shows up in the latest increase. In the previous period the shop spent €12,500 and generated €86,250 of revenue, which means the last €2,500 of budget added only €7,500. This tranche has an MER of 3.0, well below the threshold.

The average mixes money of very different value. The first thousands of budget go to brand terms, remarketing and traffic that would largely have reached the shop without advertising anyway; this is the cheapest and most profitable money. The last thousands buy more expensive clicks, broader queries and colder audiences, and in the average they are hidden behind the results of the first.

The agency’s proposal converted into profit. Revenue gains from successive budget increases are shrinking (€15,000, €11,250, €7,500), so the next €2,500 will realistically bring in about €5,000: at €17,500 the shop will generate €98,750, and average MER will fall to 5.64, still above the 5.0 threshold. Profit after advertising (€19,750 of contribution margin minus €17,500 of spend) will, however, be €2,250 compared with €5,000 at a budget of €10,000.

Before Q4 this calculation deteriorates faster. At the peak of the season, cost-per-click rises because more shops bid for the same customers, so each additional budget level buys less revenue than in spring. The incremental curve flattens earlier, and average MER shows this with a delay.

Scope of this guide. The target average MER and an annual budget derived from a profit goal are covered in a separate guide on how much an online shop should spend on marketing. Here the focus is on the profitability of each additional euro when the spend level changes: the formula, a calculation across four budget levels and decision thresholds.

Average MER answers the question of whether the budget as a whole is profitable. Whether to add another €250 is decided solely by the increment: how much shop revenue the last change in spend brought in.

Formula and threshold

Incremental MER: formula and contribution margin threshold

A budget decision needs only two figures from the shop’s history and one from its costing. The history provides revenue and total spend in two periods with different budget levels; the costing provides the contribution margin. First the definitions, then the calculation using a photography shop’s figures.

A threshold of 5.0 means that every additional €250 must bring in at least €1,250 of shop revenue. €1,250 of revenue leaves €250 of contribution margin (1,250 × 0.20), exactly what the tranche cost. Every incremental MER above 5.0 adds profit; every one below it takes profit away.

The profit or loss from a single tranche follows a simple formula: €250 × (incremental MER × 0.20 − 1). For three successive budget increases at the photography shop it looks like this:

The same threshold appears for an individual campaign, but it measures something different. The ROAS threshold calculated as 100 / margin% (described in the guide “What is a good ROAS”) applies to a single campaign and the conversion value attributed to it by the ad platform or GA4. Incremental MER works on the shop’s actual revenue and total spend, so it also covers sales that would have happened without advertising, which the platforms are happy to claim as their own.

The calculation requires data from a single place, collected day by day. Revenue comes from the shop, spend from four ad platforms (Google Ads, Meta, TikTok, Microsoft Ads), margin from purchase costs in the catalogue; stitching them together manually every month quickly leads to abandoning the measurement. Why it is worth automating is explained in the guide on automated e-commerce analytics.

You calculate the threshold for adding budget from your own contribution margin: at 20% it is 5.0, at 25% it would fall to 4.0, at 15% it would rise to 6.7. Someone else’s industry average MER does not know your cost of goods or shipping costs, so it cannot set the limit beyond which the next tranche starts reducing profit.

Calculation

Incremental MER calculation across four budget levels

A photography shop raised its budget by €2,500 every 28 days, from €7,500 to €15,000 a month. Each period covers four full weeks (P1: 3–30 March, P2: 31 March–27 April, P3: 28 April–25 May, P4: 26 May–22 June), so each has the same number of Mondays and Saturdays. Average order value held at €187.50 and contribution margin at 20%.

Average MER falls gently: from 8.0 to 6.25, and even at €17,500 it stays at 5.64, i.e. above the 5.0 threshold. Incremental MER drops much faster: 6.0, then 4.5, 3.0 and a forecast 2.0. Profit after advertising grows only with the first top-up (from €4,500 to €5,000), then shrinks to €3,750 and, according to the forecast, to €2,250.

The increment staircase shows this at a glance. Each step is another €2,500 of budget, and its height is the revenue that tranche added in the shop; below each step is the profit from every €250 of that tranche. The dashed line at €12,500 is the threshold: this is how much €2,500 must bring in at a 20% margin to break even.

The threshold crossover lies between two steps. The incremental MER from the €7.5k→€10k segment describes the euros spent at its midpoint, i.e. at about €8,750; likewise, the value of 4.5 from the €10k→€12.5k segment applies to about €11,250. Linear interpolation between these points is enough.

Maximum profit after advertising lies at about €10,000–10,500 per month. Up to this level, every top-up returns more margin than it costs; above it, each further €250 lowers profit, even though average MER stays above 5.0 the whole time. At €15,000 the shop spends about €4,500 per month beyond the crossover point and gives up about €1,250 of profit for it.

Where these figures come from: orders and revenue are a daily shop aggregate summed into periods, and spend is the sum of the Google Ads, Meta, TikTok and Microsoft Ads cost tables for the same days. The split of spend across platforms was constant (in P4: €9,000, €3,750, €1,500 and €750, €15,000 in total) and only shows where the money went. The increment calculation therefore works on the totals: total spend set against total shop revenue.

Decision

Decision thresholds: increase, hold or cut the budget

Incremental MER on its own decides nothing yet; what decides is its position relative to the threshold. Measuring the increment from two periods always carries noise: a different mix of days, a few larger lens orders, purchases of expensive equipment shifted by a week. So rather than a single line at 5.0, it is better to work with three zones and a buffer.

A 20% buffer above the threshold protects against adding budget on noise. With a 20% contribution margin, the threshold is 5.0 and the upper limit of the “hold” zone is 5.0 × 1.2 = 6.0. Only an increment of 6.0 or more gives confidence that the next tranche will make money even after typical fluctuations (in the example, +€50 for every €250).

In the photography shop, the needle sits in the “cut” zone. The last tranche (€12.5k→€15k) gave an incremental MER of 3.0, the previous one (€10k→€12.5k) 4.5. The last step above the threshold was the €7.5k→€10k jump with a result of 6.0, so that is the point to return to.

Step 4 of the procedure uses one metric from the Google Ads interface. It is worth knowing before calculating, because it indicates whether the campaigns have anywhere to spend additional money at all.

The five-step measurement procedure (using the photography shop’s figures):

1. Choose periods with different spend levels

   The comparison uses periods made up of full weeks, from 14 to 28 days, with the same number of each weekday. In the example these are four 28-day periods (3–30 March, 31 March–27 April, 28 April–25 May, 26 May–22 June) with total spend of €7,500, €10,000, €12,500 and €15,000.

2. Sum shop revenue and total ad spend per day

   Revenue comes from the shop; spend is the total across all advertising platforms, day by day. Between adjacent levels: (€86,250 − €75,000) / (€12,500 − €10,000) = €11,250 / €2,500 = 4.5.

3. Convert the result into profit from a €250 tranche

   250 × (4.5 × 0.20 − 1) = −€25. The other steps: 6.0 gives +€50, 3.0 gives −€100; a minus sign means the tranche reduces profit after advertising.

4. Check lost impression share

   In Search campaigns (Google Ads interface data) it fell from 31% at €7,500 through 14% and 3% to 1% at €15,000. At 1%, extra money only buys more expensive clicks and broader queries.

5. Run a step test

   The budget changes by 10–15% at a time and holds each new level for at least 14 days before the next step; with expensive equipment 28 days is safer. From €10,500, the next step up is approx. €11,500–€12,000, assessed with the same procedure.

Verdict for the photography shop: drop from €15,000 to around €10,500 per month. At this level, interpolated shop revenue is approx. €77,250, contribution margin approx. €15,450, and profit after advertising approx. €5,000 compared with €3,750 today. The agency’s proposal (€17,500, forecast profit €2,250) is best rejected until there is evidence that Q4 shifts the incremental curve upwards.

The cut of approx. €4,500 starts where the platform-reported increase in conversion value is weakest. At €15,000, spend according to the platforms splits into Google Ads €9,000, Meta €3,750, TikTok €1,500 and Microsoft Ads €750; in each platform you compare how much the reported conversion value grew between the €12,500 and €15,000 levels relative to the increase in spend. This is a platform estimate that sets the order of cuts; the effect of the cut itself is assessed again on shop revenue and MER.

Pitfalls

Five pitfalls when calculating budget increments

Incremental MER compares two periods, so it measures everything that changed between them. Budget is only one of those changes; season, prices, stock availability and window length can shift the result more than an extra €2,500. Before the figure feeds into a decision, run through five checks.

1. Seasonality adds itself to the budget effect. In photography, May and June have naturally higher demand (weddings, first communions, trips), so part of the increase between P2 and P3 would have come without any extra spend. The year-on-year index for the same weeks is 1.03, so the expected P3 revenue without a budget change is €75,000 × 1.03 = €77,250; the budget is left with €86,250 − €77,250 = €9,000 of increment and an incremental MER of 3.6 instead of 4.5.
2. Promotions, price changes and stock-outs distort the comparison. If a lens sale ran in one of the periods, or a popular camera body was out of stock, revenue changed for reasons unrelated to the budget. Each such day should be marked in the shop’s event calendar, and periods with a similar background compared, or the disrupted days excluded from both periods.
3. Platform revenue overstates the effect of extra spend. Google Ads, Meta, TikTok and Microsoft Ads each claim parts of the same sales, so the sum of platform-reported increases in conversion value comes out higher than the increase in actual takings. The numerator of incremental MER is shop revenue only; platforms serve to estimate where the money went.
4. Too short a window for expensive equipment. With an average basket of €187.50, the decision to buy a camera or lens takes weeks, so spend rises from day one while revenue catches up with a delay. A weekly comparison understates the increment after an increase and distorts the reading after a cut (sales from earlier weeks are still coming in), so for large budget changes the window is at least 28 days of full weeks.
5. Q4 shifts the entire incremental curve. Higher demand raises the increment at the same budget levels, but rising cost-per-click lowers it, so a summer measurement does not carry over directly to November. Budget levels for the peak are compared with last year’s Q4 (the same incremental calculation on those weeks), and only that result goes into the Q4 season plan.

Seasonal adjustment changes the assessment of the P2→P3 tranche (illustrative data, 20% contribution margin):

The direction of the decision stays the same, as both values are below the 5.0 threshold. However, the loss on the tranche grows almost threefold: from €25 to €70 per every €250 (250 × (3.6 × 0.20 − 1) = −€70). Without the adjustment, the P2→P3 tranche would look like “almost break-even” spend, even though it clearly reduced profit.

DataOrganizer

How DataOrganizer calculates incremental MER

Incremental MER requires four data series on the same daily cadence: shop revenue and orders, spend from all advertising platforms, margin from product cost, and impression share lost due to budget. DataOrganizer holds them in a single model. Shop orders and revenue arrive as a daily aggregate, costs come from Google Ads, Meta, TikTok, Microsoft Ads and the other connected platforms, margin from the product_cost field, and lost share (search_budget_lost_impression_share) from Google Ads campaigns.

Data refreshes every morning with the complete previous day. The 28-day periods therefore consist of full days, with no exports from dashboards or stitching of spreadsheets, and each new day of a budget test is added to the comparison automatically. The incremental question looks like this in the app:

Shop revenue remains a single figure per period in this calculation. If the assistant shows where the money went (Google Ads 60%, Meta 25%, TikTok 10%, Microsoft Ads 5%), it does so based on the platform dashboards and labels that split as an estimate. You can ask the same question in Claude or ChatGPT via DataOrganizer MCP if that is where you work day to day.

Methodological note: the figures in the examples are illustrative. Variable costs (payments, shipping, packaging) are provided by the shop, as they do not follow from product_cost. Incremental MER describes the correlation between budget level and shop revenue across successive periods, so it requires seasonal adjustment and stable conditions: no major promotions, price changes or stock-outs in the compared windows.

Methodological note: the figures in the photography shop example are illustrative, and variable costs (payments, shipping, packaging) are provided by the shop, so the contribution margin must be calculated on your own data. Incremental MER describes the correlation between budget level and shop revenue. It requires seasonal adjustment and stable conditions (no promotions, price changes or stock-outs in the compared periods), and the split of spend across platforms comes from the dashboards and is an estimate.

FAQ

Common questions about increasing the ad budget

When should an online shop increase its ad budget?

When the last budget increase brought in shop revenue at least 1 / contribution margin times its cost. At a 20% margin, the threshold is 5.0, so the next €250 must generate at least €1,250 in revenue; in practice, budget is added at an incremental MER of 6.0 or more, held at 5.0–6.0 and cut below 5.0. A supporting signal is the impression share lost due to budget in the Google Ads dashboard: when it falls to a few per cent, extra money mainly buys more expensive clicks.

How do you calculate incremental MER?

Incremental MER = (shop revenue in period B − revenue in period A) / (total spend B − total spend A), for periods made of full weeks of equal length. Example: spend rises from €12,500 to €15,000 and revenue from €86,250 to €93,750, so 7,500 / 2,500 = 3.0. The profit on every €250 of this tranche is €250 × (3.0 × 0.20 − 1) = −€100; revenue is taken from the shop, and spend is summed across all platforms.

How does incremental MER differ from average MER and ROAS?

Average MER divides the shop’s total revenue by total spend and answers whether the budget as a whole is profitable; incremental MER measures only the effect of the most recent change in spend level. ROAS is calculated from conversion value in the ad platform or GA4 for a single campaign, and its threshold is covered in the guide “What is a good ROAS”. In the example, the average MER of 6.25 is above the 5.0 threshold, while the last tranche reaches only 3.0.

How long should you test a new ad budget level?

At least 14 days at one level, and for expensive products such as cameras and lenses, 28 days, because the purchase decision takes weeks and revenue catches up with spend after a delay. Change the budget in steps of 10–15%, compare full weeks with the same number of each weekday, and make sure the comparison period contains no promotions, price changes or stock shortages.

Is it worth raising the ad budget before Q4?

Only if last year’s Q4 shows that incremental MER stayed above the margin-based threshold at higher budget levels, because higher demand raises the increment while rising cost per click lowers it. In the example, the proposed €17,500 gives an incremental MER of 2.0 and a profit of about €2,250 versus €5,000 at €10,000, so without such evidence a stepped test is the more sensible option. DataOrganizer combines shop revenue and spend from all platforms day by day, including last year’s season described in the Q4 plan.