A Revenue Jump Might Be a Price Hike, Not More Customers

A case study puts two revenue figures side by side, before the campaign and after, and the gap between them is the argument. Both numbers can be accurate and the chart honestly drawn, and the claim beside it still wrong, because revenue can rise without a single extra customer buying anything.

The case study that reads cleaner than it is

Picture a hypothetical case study, with hypothetical numbers, for a specialty homewares retailer. Year 1 revenue is 4 million, Year 2 revenue is 5 million, and the write up reports 25 percent year over year growth, credited to a twelve month social campaign across Year 2. Now add one hypothetical fact it does not mention: in month seven of Year 2, the retailer put a 10 percent list price increase across its catalogue. Not a promotion, a permanent change inside the campaign period.

Read the case study as given and all 25 points belong to the campaign. It never says whether the retailer sold more in Year 2 than in Year 1, and revenue alone cannot answer that. That is an external factor lining up with the campaign window, except the factor is the business’s own pricing.

Revenue is two numbers wearing one number’s clothes

Revenue is average price multiplied by units sold. Either factor can move the total alone. Selling 20 percent more units at the same price and selling the same units at prices 20 percent higher both produce 20 percent revenue growth: same chart, same headline, different claims about what the campaign did.

Official statistics handle this split. The US Bureau of Economic Analysis defines nominal or current dollar estimates as figures valued in the prices of the period when the transactions occurred, and quantity or real estimates as figures that exclude the effects of price changes. A case study reporting revenue only publishes the first and invites you to read the second.

What the case study would need to show before the campaign gets full credit

Four figures would settle it, and any business with an order system has them.

• Unit or order count for both years. If the count is flat while revenue is up 25 percent, the campaign brought in no demand.
• The size, timing and type of any price change. A list price rise, a standing discount withdrawn and a plan tier repriced are three different events with the same effect on the total. Which one, how big, and when relative to the campaign.
• Average order value or average price per unit, both years. This shows whether customers pay more per transaction. If it rose by roughly the price change while unit count held, the story is arithmetic.
• Who the price change applied to. New customers, existing customers, or everyone. A rise hitting only new customers entangles pricing with acquisition, so volume has to be split by cohort.

Working the hypothetical through

Back to the hypothetical retailer and its hypothetical figures: 4 million, then 5 million, 25 percent reported, a 10 percent rise halfway through Year 2. Because the increase covered only the second half, the average price across Year 2 was roughly 5 percent above Year 1, not 10. Divide Year 2 revenue by 1.05 and you get about 4.76 million, what that year would have earned at Year 1 prices.

Set that hypothetical 4.76 million against the 4 million baseline and volume driven growth is about 19 percent, leaving roughly six points of the reported 25 to the price change. The headline is not wrong, it is compound, and the campaign’s share is now a smaller, more defensible number.

It is an estimate: it assumes sales spread evenly across the year and nobody buying less at the higher price. The point is not precision, it is that the split is big enough to matter, a rough stand in for the counterfactual a case study can’t show you.

Where this same question hides outside a simple price increase

The subtler versions sit inside subscription and plan structures. Repricing an existing tier lifts revenue per customer with no new customers, and so does a push that moves existing accounts up a tier: customer count flat, revenue line climbing. Bundling works the other way round: raise the free shipping threshold and customers spend more per transaction without buying more often.

One thing this is not: a currency effect. If a business reports in one currency and sells in another, an exchange rate move can inflate the total with nothing changing underneath. The concern here is a deliberate pricing decision inside the reporting window, and the two are worth keeping apart.

The one question that closes the gap

For any multi year revenue comparison, one question does most of the work: what happened to unit or order volume over the same period, and did price change in between? A two year window is exactly where a mid period price change sits unnoticed, which is why the attribution window itself is a choice and why the ‘before’ point in a comparison is also a choice.

Some write ups answer it before you ask. One that puts order count or average order value next to revenue, and states pricing held across both periods, has closed the gap itself. Where those figures are missing, the absence is informative without being damning: it may simply be that nobody thought to include them.

See how the disclosure varies

The way to build an eye for this is to read results write ups back to back and notice which ones state volume alongside revenue. Browse the case study library, every result credited to the agency that published it, or submit your own.

FAQ

If a case study does not mention a price change at all, should I assume there wasn’t one?

No. Absence of disclosure is not evidence of absence, and a write up that says nothing about pricing may simply not treat pricing as reportable. The volume question is unresolved: you do not know prices moved, and you do not know they held. Treat the growth figure as an upper bound.

Is a percentage growth rate ever safe to trust without a volume figure?

The condition that would satisfy the question is narrow: the write up states unit or customer counts for both periods and confirms pricing was unchanged. With both present, the percentage does what the reader assumes it does. That is the bar.

Sources

• US Bureau of Economic Analysis, Glossary: Quantity estimates, fetched 6 September 2026.
• US Bureau of Economic Analysis, Glossary: Current-dollar estimates, fetched 6 September 2026.