The Challenge
A catalogue spanning pet harnesses, board games, power banks, landline phones and rocking chairs has no single audience, no shared keyword set and no consistent margin. Judged on ROAS alone the advertising looks weak — and every previous review of the account had reached exactly that conclusion, then cut budget on campaigns that were doing useful work.
Quick answer
A multi-category seller on Amazon India — pet supplies, toys, consumer electronics, small furniture — returned 2.91× ROAS at 34.32% ACoS across 90 days. That is the weakest advertising return in our managed book.
It is also an account producing ₹1.92 Cr of revenue at 8.07% TACoS, with 76.5% arriving organically. Both statements describe the same business, and only one of them describes its health.
The numbers
| Metric | Value |
|---|---|
| Total Amazon revenue | ₹1.92 Cr |
| Ad-attributed sales | ₹45.20 L |
| Ad spend | ₹15.51 L |
| ROAS | 2.91× |
| ACoS | 34.32% |
| TACoS | 8.07% |
| Organic share | 76.5% |
| Total orders | 19,399 |
| Active campaigns | 156 |
Period: 4 June – 1 September 2026, settled through 31 August. Live account figures.
Why the name is withheld
Client name protected under NDA. All metrics from the live account.
The point of publishing this one
Put this account next to the home décor brand elsewhere in these case studies:
| This account | Home décor brand | |
|---|---|---|
| ROAS | 2.91× | 5.60× |
| ACoS | 34.32% | 17.86% |
| Organic share | 76.5% | 34.7% |
On advertising metrics, the décor account wins comfortably. On business health, this one does. Three-quarters of its revenue would survive if advertising stopped tomorrow; barely a third of the décor brand’s would.
ROAS measures the efficiency of the advertising. It says nothing about whether the business underneath needs the advertising. Choosing an agency on ROAS alone selects for the first and ignores the second.
Why ROAS is structurally low here
Nothing shares a keyword. A pet harness and a landline phone have no overlapping search terms, no shared audience and no common bidding logic. Each category has to be prospected independently, so a large share of spend is always working on terms that are still being learned rather than terms already proven.
Margins differ by category. A 34% ACoS is comfortable on a ₹2,148 vest harness and unacceptable on a ₹388 toy car. A single account-level ACoS target across that spread guarantees over-spending on the cheap items and under-spending on the profitable ones — so we do not set one.
Discovery costs money in every category simultaneously. 156 active campaigns against 714 paused: most of what gets tested is turned off. That testing shows up as spend without return, which is exactly what it is, and it is what produced the 76.5% organic base.
What we do differently on accounts like this
Targets are set per category, against that category’s margin — not once for the account. Campaigns are judged on contribution after product margin, not on a shared ACoS line. And the metric on the front page of the monthly report is TACoS with organic share beside it, because that pair describes the business while ROAS describes only the ads.
Is a 2.91× ROAS acceptable?
On a multi-category catalogue with a 76.5% organic base and 8.07% TACoS — yes, and cutting spend to raise it would likely shrink the organic base that makes the account strong. On a single-category brand with 35% organic, the same 2.91× would be a serious problem. The number cannot be judged without the context, which is why we publish both.
Free Amazon audit — we will run the same diagnostic on your account.
Result
2.91× ROAS at 34.32% ACoS — the lowest return in our managed book — on an account where 76.5% of ₹1.92 Cr revenue arrives organically at 8.07% TACoS.
₹1.92 Cr
Total Amazon Revenue
2.91×
Blended ROAS
34.32%
ACoS
8.07%
TACoS
76.5%
Organic Share
90 days, Jun–Aug 2026
Period
Under NDA
Client Name