Shelf Share vs Share of Shelf: What FMCG Brands Are Actually Measuring (And Getting Wrong)
A brand manager at a mid-sized snacks company in Dubai once sent me a WhatsApp at 11pm. He was arguing with his agency about whether their shelf share had gone up 12% or down 3%. Same stores. Same month. Same photos. Two completely different numbers.
The reason? They were measuring two different things and calling them the same word.
This happens constantly. And honestly, I used to conflate the terms myself in early sales decks before someone at a Riyadh distributor politely corrected me over shawarma. So let's sort it out properly, because if you can't define what you're measuring, you can't defend your trade spend to the CFO.
The actual difference (and why nobody agrees on it)
Here's the thing — there's no ISO standard for this stuff. Different brands, agencies, and platforms use these terms differently. But the working definition most FMCG operators I talk to land on goes like this:
Share of shelf is your brand's linear space divided by total category linear space. If the biscuit aisle has 400cm of shelf and your brand occupies 68cm, your share of shelf is 17%. It's a physical, measurable, boring number. And it's the one that correlates most tightly with offtake in almost every study I've seen.
Shelf share (as most people use it) is broader. It can mean facings count, SKU count, vertical position weighting, eye-level premium, or some blended visibility score. Some agencies calculate it as facings × visibility multiplier. Some just count SKUs on shelf regardless of size.
So when your agency says your shelf share went up 12%, ask them: facings, linear cm, SKU count, or a weighted score? Because those four things move independently. You can gain 3 facings and lose 40cm of linear space if a competitor swaps in bigger packs.
The snacks brand I mentioned? Agency was counting SKUs on shelf (went up because they added a new 15g pack). Brand manager was measuring linear cm (went down because the 150g family pack got delisted in 4 stores). Both right. Both useless without context.
What you should actually measure
Look, if I had to pick one metric for an FMCG brand starting from scratch, I'd pick linear share of shelf in centimeters, measured at the category level, per store cluster. Here's why.
Facings lie. A single facing of a 2-litre bottle takes 4x the shelf real estate of a 500ml. Counting facings punishes big-pack strategies and rewards proliferation of small SKUs, which isn't always what you want.
SKU count is even worse. It tells you presence, not visibility. Your competitor could have 6 SKUs at ankle height and you'd have 4 at eye level and their SKU count would look better on paper.
Linear cm is honest. It's what the shopper's eye actually sweeps across. Pair it with vertical position (eye level, mid, bottom) and you've got 80% of what matters for in-store visibility metrics.
The other things worth tracking, in rough order of importance:
- Out-of-stock rate by SKU by store. Because a 22% share of shelf FMCG number means nothing if the top-selling SKU is empty.
- Planogram compliance. Are you where the retailer said you'd be?
- Price compliance. Is your MRP visible and correct?
- Adjacency. What's next to you? Being next to the category leader is often better than being next to your own brand extension.
- Promotional visibility. POSM, wobblers, gondola ends — are they up when they're supposed to be?
How to actually measure it without losing your mind
This is where most brands fall apart. They agree on the metric, then they can't collect the data reliably. A rep with a clipboard measuring 40 stores a week is not going to give you clean shelf share measurement data. I've watched it fail in Karachi, in Manchester, in Muscat. Same story everywhere.
What's working now, and this is genuinely why we built the shelf photo analysis piece into Zivni, is image-based measurement. Rep takes a photo of the shelf. AI identifies SKUs, measures linear space, flags out-of-stocks, calculates share of shelf against the category. Takes about 8 seconds per shelf. The rep doesn't argue about numbers because they didn't produce them.
But — and this matters — the AI has to be trained on your specific categories and packaging. Generic models struggle with private label lookalikes, with new pack designs, with the weird lighting in a hypermarket freezer aisle. If you're evaluating any platform (ours or otherwise), ask for a pilot on 20 of your actual stores before you sign anything. If they won't do that, walk away.
Also: measure the same stores at roughly the same time of week. Shelf share on a Monday morning after weekend restocking looks very different from Friday evening. I've seen brands celebrate a 4-point gain that was really just a timing artifact.
What the number should trigger
A share of shelf report that just sits in a PowerPoint is worse than useless. It's expensive theater. The number has to trigger a specific action.
For us, a 5-point drop in a store cluster triggers a rep visit within 48 hours, a photo review, and either a merchandiser deployment or a conversation with the retailer's category manager. A 5-point gain triggers a case study we send to other retailers as social proof. Everything else is noise.
And if your shelf share is climbing but sell-through isn't, you have a different problem — probably pricing, probably promotion timing, possibly a category that's just shrinking. Don't confuse visibility with velocity. I've watched brands double their facings and lose money because they were winning shelf space in stores that were losing shoppers.
Which brings up the real question — do you know your top 20 stores by absolute contribution, and do you know your share of shelf in each of them right now, this week? Because if you don't, that's where I'd start on Monday.