Single-Product Ad Budget: Where the Next Yen Goes, Not % of Sales

I run a small store that sells a single product, and every month I get stuck on "how much ad budget to put in which channel." The usual articles tell me "ad spend should be 20-30% of sales," but that is about how much to spend overall — it says nothing about which channel the next ¥10,000 should go to. With one product, I can't spread sales across a product mix, so the precision of this channel choice decides, directly, the profit I keep. This article lays out — as plainly as I can — how to allocate budget by real per-channel profit and by "how much the added spend actually returned," rather than a total-spend rule of thumb.

TL;DR#

1. "What percentage of sales" is only a total-budget guideline

   It answers how much to spend overall — not which channel the next ¥10,000 belongs to

2. Decide the next ¥10,000 at the margin

   Look at the extra revenue the added spend produced, not the average ROAS. Topping up a saturated channel won't grow it

3. Allocate by real per-channel profit

   Not the platforms' self-reported numbers — align your own revenue on one common basis and compare it by RPS and contribution profit

4. One product means there's nowhere to hide

   You can't spread sales across a product mix, so channel precision converts straight into profit

1. Why a percentage of sales cannot allocate budget#

Bottom line: "percentage of sales" is a total-spend guideline; it does not tell you which channel the next ¥10,000 belongs to.

Japan's BtoC physical-goods EC market keeps growing[4], and within it, single-product commerce — betting on one item — is often quoted at "ad spend of 20-30% of sales," varying by category[1]. For subscription cosmetics and supplements it can exceed 30%[3]. As a rough ceiling on how much you may spend on ads overall, that is useful.

But the decision you actually agonize over daily is elsewhere: "how do I split the budget I have across search ads, social ads, and retargeting?" and "if I can add ¥10,000 next month, which channel returns the most?" This allocation — the next move — is the judgment that drives profit. A total-spend guideline like "percentage of sales" answers none of it. Even the KPIs you should watch to recover ad spend in single-product subscription commerce lean toward per-channel efficiency, not a total ratio[2].

And a single-product store sells exactly one item. A multi-product store can cover a missed ad on one item with another; a one-SKU store cannot. Get the channel split wrong and the loss hits the whole business directly. That is why you can't rest easy on a total guideline — you have to look one level deeper into the allocation itself.

2. Place the next ¥10,000 at the margin#

Bottom line: judge a channel not by its average ROAS but by the extra revenue the added spend produced — the margin.

The most important question in allocation is "where does the next ¥10,000 go?" What you should look at here is not the channel's average ROAS (return on ad spend). It is how much extra revenue the added spend generated. We call this "looking at the margin."

Say a channel takes ¥100,000 a month and returns ¥300,000 — an average ROAS of 3x. Add ¥10,000, though, and revenue may not rise by ¥30,000. If you've nearly reached everyone likely to buy, that extra ¥10,000 might add only ¥5,000. That is channel "saturation."

So the next ¥10,000 belongs not in "the channel with the highest average ROAS" but in "the channel that still grows cleanly when you add to it." A high-average-ROAS channel that's already saturated wastes the top-up. A plain-looking channel with headroom returns more revenue on the same ¥10,000. For why more ad spend doesn't always mean more profit, see When more ads stop paying: spotting ad saturation.

3. Compare channels by real profit: RPS and contribution#

Bottom line: allocate on your own revenue aligned across channels — compared by RPS and contribution profit — not the platforms' self-reports. The idea is simple; the heavy part is re-aligning every channel to the same basis each month.

Even once you decide to judge at the margin, you can't if the numbers behind it are scattered across platforms. The ROAS each ad platform reports is a self-report calculated to suit that platform, and the same sale is often double-counted by several platforms claiming it as their own. So before allocating, you have to align your own real revenue on one common basis across all channels. The gap between platform self-reported ROAS and the real efficiency seen across the whole site (MER) is laid out in Platform ROAS vs MER: why they differ.

A handy tool here is RPS (revenue per session) — channel revenue divided by that channel's sessions, i.e., the average revenue one visit produced. The channel with the most traffic isn't the winner; a channel with few visits that convert well is more efficient — that shows up in this single number. How to compute RPS and where to use it is collected in RPS (revenue per session): the complete guide.

RPS          = channel revenue / channel sessions
Contribution = revenue - (cost + shipping + payment fees + ad spend)

Single-product commerce has one advantage here. With one product, cost, shipping, and payment fees are roughly fixed — so contribution profit (revenue minus the costs directly incurred, i.e., the profit you keep) is easy to estimate. A multi-product store is fiddly because each item has a different cost; with one SKU the "revenue ranking" is close to the "profit ranking."

The idea itself isn't hard. You align revenue and sessions by channel, compute RPS, add an approximate contribution profit, and compare. What's hard is doing this every month. Smoothing out the double-counting in each platform's self-reported ROAS, reconciling channel names that are written inconsistently, excluding automated programs (bots), and then continuing to line up every channel on the same basis — keeping this preprocessing alive by hand each month is quietly heavy work. And a view that puts each channel's "RPS × real revenue" on one screen to compare simply isn't there, structurally, in GA4's standard reports. So RPS and contribution profit are a "starting point for narrowing down where the next ¥10,000 goes" — but not something that, by themselves, settles every month's allocation.

One more thing: when you evaluate channels, also checking "is it bringing in new customers" sharpens the call. For the trap where a high-ROAS platform is actually skewed toward returning customers, see Are high-ROAS channels the risky ones? Evaluating channels by splitting new from returning.

RevenueScope solution

Bottom line: to decide "where the next ¥10,000 goes" by margin and real profit, you have to align each platform's self-report onto one basis and lay every channel's ad spend and revenue side by side on a single screen. RevenueScope provides the material for that reconciliation.

Even when you want to decide "where the next ¥10,000 goes" by margin and real profit, each platform's self-reported ROAS is inconsistent — and "across all channels combined, how much more can I still spend efficiently?" never surfaces from looking at ad dashboards one at a time. That is where single-product budget allocation stumbles most.

RevenueScope takes each platform's ad spend by form entry or CSV upload and reconciles it against your own revenue tracking. That lays each channel's ad spend beside the revenue of the matching channel. From there, you estimate by hand the true ROAS (return on ad spend) and the saturation — how much efficiency has already dropped by the time you add one more yen. RevenueScope displays measured ROAS only for platforms whose traffic is almost entirely ads, such as Google Ads and Criteo. Where organic revenue falls into the same channels, as with Meta, it shows ad spend and revenue side by side and no ROAS. Saturation itself does not appear on screen.

For example, RevenueScope lets you see each channel's ad spend and the revenue of the matching channel side by side in one list. Below is a worked example of computing true ROAS by hand from those two figures.

An example with figures for a fictional site. It is not a display of the RevenueScope screen.

What works here is reading ROAS together with how saturated a channel is. Even a high-ROAS channel won't grow cleanly on the next ¥10,000 if it is near its saturation ceiling. Conversely, a channel with plain ROAS but headroom before saturation still returns on the same ¥10,000. The "next ¥10,000" belongs not in the highest-average-ROAS channel but in the one with headroom before saturation. You estimate the degree of saturation by hand, from how much revenue rose in the months you raised ad spend.

We tend to assume "raise the budget and sales rise too," but if current channels are saturated, most of the added spend only lowers efficiency. "How much can you still add efficiently to current channels" and "from where should you shift to opening new channels or refreshing creative" are calls you make yourself, by laying every channel's ad spend and revenue on one screen and watching how much each increase returned.

What RevenueScope lays out is, for channels whose ad spend you have entered, the ad spend and the matching revenue, plus per-channel revenue efficiency (RPS). From this you build your own first-draft allocation. Focused on these revenue-based materials, it narrows down which channel the next ¥10,000 belongs to. Contribution profit and gross margin are the last step you finish on your own side by entering your business's cost, shipping, and fees — and because RevenueScope shows you the revenue ranking, laying the profit ranking on top of it is a light final touch. What RevenueScope takes over is the heavy manual work of re-aligning this "where the next ¥10,000 goes" across every channel, every month, so you can focus on the final allocation call.

Even at the stage where you haven't entered ad spend yet, RevenueScope lists high- and low-efficiency channels by RPS (revenue per session). Enter your ad spend and you can lay ad spend beside the matching revenue and build the allocation draft yourself: a two-tier setup. The more a business rides on a single product, the more this "where the next ¥10,000 goes" precision decides profit directly.

FAQ#

Frequently asked questions#

Q. So how much should a single-product store actually spend on ads?

A. There's no single right number. "20-30% of sales" is a rough ceiling on the total, but what matters is how you split budget within that cap. Before worrying about the total, comparing per-channel efficiency (RPS) and checking whether budget is flowing to the channels that still grow has a bigger effect on profit.

Q. Should I prioritize search or social?

A. Deciding by your own numbers is surer than deciding by channel type. Search ads generally reach "people ready to buy now," so they're said to work early — but if already saturated, topping up is wasted. Look at each channel's RPS and the lift from adding ¥10,000 (the margin), and put money where it still grows.

Q. Do I have to calculate contribution profit precisely?

A. A rough figure is fine at first. With one product, cost, shipping, and fees are roughly fixed, so even a rough estimate keeps the revenue ranking close to the profit ranking. Start by comparing revenue and RPS by channel, then add an approximate contribution profit. If the ranking changes from the ROAS view, that's proof ROAS-only judgment is risky.

Conclusion#

A single-product store's ad budget can't be allocated by "what percentage of sales." That's a total-spend question; it doesn't tell you which channel the next ¥10,000 belongs to. Allocate by the margin — "how much the added spend actually returned" — not average ROAS, and compare channels by real per-channel profit (RPS and contribution profit) rather than the platforms' self-reports. Precisely because there's one product, cost is roughly fixed so profit is visible, and there's nowhere to hide a bad channel choice.

But even when the idea is simple, what actually works is "re-aligning every channel to the same basis, every month." Only after you smooth out the per-platform double-counting, exclude bots, and put each channel's ad spend, revenue and RPS on one screen does "where the next ¥10,000 goes / from where to shift to a different move" become decision material. Keeping this cross-channel re-alignment going by hand each month is the heaviest part of single-product budget allocation. RevenueScope takes over the part of this heavy repetition where each channel's ad spend, revenue and RPS are lined up, so you reach a state where the next ¥10,000's destination can be weighed with real numbers.

References#

• [1] Tsuhan Tsushin, "What percentage of sales is the right ad-spend level for mail-order businesses? Category averages and rankings," 2024
• [2] Niche Media, "What is single-product subscription commerce? The KPIs needed to recover ad spend and make the model work," 2021
• [3] Netshop Manager Forum, "Ad tactics for single-product commerce sites, part 2: goal-setting and using video to maximize sales," 2022
• [4] Ministry of Economy, Trade and Industry, "FY2024 Survey on E-Commerce Market," 2025