How to Calculate ROI in FMCG: A Practical Guide with Real Examples

By Sufyan · 2026-08-25 · 5 min read

Last month a distributor in Karachi showed me his monthly settlement sheet. Three trade promotions running at once. Two shopper campaigns. A 12% bonus scheme on a slow-moving SKU. He was proud of the volume growth — 23.4% quarter on quarter — until I asked him the obvious question. "What's your ROI on the bonus scheme?"

He didn't know. Nobody in the room knew.

This is the FMCG story I hear over and over, from Lagos to Lahore to Jakarta. Sales teams celebrate volume. Finance quietly cries. And somewhere in between, millions in trade spend go unmeasured because the math feels harder than it actually is.

So here's how to actually calculate ROI in FMCG — the way I've seen it work at brands that survived, and the mistakes I made when I tried it the first time.

The formula everyone forgets to adjust

Basic ROI is simple: (Return − Investment) / Investment × 100.

But FMCG breaks this formula in three ways, and if you don't adjust, your numbers will lie to you.

First, you can't count gross sales as "return." You need incremental gross margin — the extra margin you earned because of the activity, not the margin you would've earned anyway. Second, trade spend isn't the only investment. You've got sampling costs, merchandiser hours, printing, logistics, and (this one hurts) the margin you gave away as discount. Third, cannibalisation. If your promoted SKU stole volume from a sister SKU in the same portfolio, that's not real growth.

The adjusted formula I use:

ROI = (Incremental Volume × Net Margin per Unit − Total Activity Cost) / Total Activity Cost × 100

Sounds obvious written out. Almost nobody does it correctly.

A distributor scheme example

Let's say you run a 10% bonus scheme (buy 10 cartons, get 1 free) on a shampoo SKU in a secondary market. Baseline monthly offtake from this distributor is 800 cartons. Your net margin per carton (after COGS, freight, and normal trade margin) is $4.20.

During the scheme month, offtake jumps to 1,240 cartons. Feels great, right?

Now do the math properly.

Incremental volume: 1,240 − 800 = 440 cartons. But the bonus itself gave away 1,240/11 ≈ 113 free cartons, costing you 113 × cost price ($9.10) = $1,028. Add merchandiser deployment ($240) and a small POSM print run ($180). Total investment: $1,448.

Incremental margin: 440 × $4.20 = $1,848.

ROI: ($1,848 − $1,448) / $1,448 × 100 = 27.6%.

Positive. But not the hero number the sales team was expecting when they saw 55% volume growth. And we haven't even touched forward-buying — the ugly reality that a chunk of those 440 "incremental" cartons will just sit in the distributor's warehouse and eat into next month's order.

Honestly, I got this wrong for years. I'd celebrate a scheme, then wonder why the following month's sell-in dropped 30%. Because I was double-counting. The distributor wasn't buying more — he was buying earlier.

Trade promotions and shopper campaigns

Trade promotions are trickier because the "baseline" is a moving target. A biscuit brand running a 15% off promotion at a modern trade chain in Dubai isn't competing against last month's sales — it's competing against what would've happened without the promo, which includes seasonality, competitor activity, and store footfall.

Best practice: use a control store or a control period. Pick a similar chain not running the promo, or use the same store's non-promoted weeks as baseline. Then apply the same adjusted ROI formula.

One real fmcg roi example from a beverage client: they ran a $18,000 in-store activation at 42 supermarkets. Uplift looked like 8,900 incremental units at a margin of $1.15 = $10,235. ROI = ($10,235 − $18,000) / $18,000 = −43%. Terrible. But the brand team argued for the "brand equity halo." Fine — but measure that separately. Don't mix a hard ROI number with a soft brand argument. That's how CFOs stop trusting marketing.

For digital campaigns, the calculation is cleaner but the attribution is messier. If you spent $6,200 on a Meta campaign that drove 1,410 units of trackable e-commerce sales at $2.80 margin per unit, your direct ROI is ($3,948 − $6,200) / $6,200 = −36%. Add offline uplift (measured via a lift study or a modern MMM) and you might get to positive. Might.

Why most FMCG teams get this wrong

Data fragmentation. That's the honest answer.

The scheme lives in a spreadsheet on the sales manager's laptop. The primary sales sit in the ERP. Secondary sales come from distributor claims, which arrive 21 days late and are usually wrong. Merchandiser hours are on paper. POSM costs sit with the marketing agency. By the time someone tries to calculate ROI, half the data is stale and the other half is estimated.

This is exactly why field sales platforms have become non-optional for serious FMCG operators. Tools like Zivni pull primary orders, secondary sell-out, merchandiser attendance, and scheme redemption into one view — which means the ROI math I described above isn't a quarterly finance exercise. It's a Tuesday morning conversation. When your area sales manager can see that Scheme A is running at 27% ROI and Scheme B at −11%, he kills Scheme B before week two instead of week eight.

A quick checklist I give any FMCG team trying to fix this:

Define baseline before the activity starts. Write it down. Sign it.
Separate incremental margin from headline sales growth
Account for forward-buying by looking at a 60-day window, not 30
Include every cost — merchandisers, POSM, freight on free goods, everything
Set a minimum ROI threshold (I use 15% for trade schemes, 25% for consumer promos) below which activities auto-review

One number that changes everything

If you only track one metric, track Return on Trade Investment (ROTI) — total incremental gross margin divided by total trade spend, calculated monthly, per distributor, per SKU cluster.

When a brand starts publishing ROTI internally, behavior changes fast. Sales managers stop asking for more scheme budget and start asking for smarter scheme design. Distributors who were consistently loss-making on trade spend get renegotiated. The marketing team stops proposing activations without a hypothesis.

And here's the thing — none of this requires a data science team. It requires one clean spreadsheet, an honest baseline, and the willingness to see a negative number and not flinch.

Which brings me back to that distributor in Karachi. We rebuilt his last six months of schemes with real numbers. Three out of eleven were positive ROI. The rest? He'd been paying to grow. Now the question is whether he wants to keep doing that.