How to Set Up a Merchandising Audit Program for FMCG: From KPIs to Field Execution

By Sufyan · 2026-06-29 · 4 min read

The first merchandising audit program I helped a brand set up in Dubai had 73 questions per outlet. Seventy-three. Reps were spending 18 minutes per store just answering checkboxes, and half the data was useless by the time it hit the dashboard.

We cut it to 12 questions. Audit completion jumped from 41% to 89% in six weeks.

That's the thing nobody tells you about merchandising audits. The program isn't about catching everything. It's about catching the right things, consistently, in a way your field team can actually execute between 9am and 6pm without hating their lives.

So here's how I'd build one from scratch today, after watching dozens of FMCG brands across the GCC and Pakistan get this wrong (and a few get it brilliantly right).

Start with the KPIs, not the checklist

Most teams do this backwards. They sit in a conference room, brainstorm everything they wish they knew about every outlet, and end up with a 60-question monster. Then they're shocked when reps skip half of it.

Work the other direction. What decisions do you actually make with audit data?

If you're a biscuit brand, you probably care about: share of shelf vs. your top two competitors, planogram compliance, OOS (out of stock) rate on your hero SKUs, secondary placement presence, and POSM visibility. Maybe price compliance if you've had MRP issues. That's it. Five or six FMCG merchandising KPIs that actually move budget decisions.

Anything else is nice-to-know. And nice-to-know kills audit programs.

Here's a quick gut check I use with clients: for each proposed KPI, ask "if this number gets worse next month, what will we do about it?" If nobody can answer in one sentence, drop the KPI. You're not running a research project. You're running a sales operation.

Build the retail merchandising audit checklist around outlet type

A hypermarket in Riyadh doesn't get audited like a kiryana store in Karachi. I see brands try to use one universal checklist and it's painful for everyone.

Segment your audit checklist by channel — modern trade, traditional trade, HoReCa, wholesale — and within each, by tier. A Carrefour gets the full planogram check with shelf photos. A tier-3 grocery store gets a 4-question version: are we in stock, is the price right, is the POSM up, are competitors doing anything weird.

For a typical modern trade audit, I'd include:

Notice what's not there. No "rate store cleanliness 1-5." No "describe the manager's mood." Subjective scoring across hundreds of reps creates noise, not insight.

One more thing on the checklist — photos are non-negotiable. Every audit needs at least one geo-tagged, timestamped shelf photo. Without that, you're trusting a tired rep at 4pm to be honest about facings. Even the best reps fudge when they're rushing.

Make field execution stupidly simple

This is where most merchandising audit programs die. The KPIs are fine. The checklist is fine. But the app is a nightmare, the SKU list takes 40 seconds to scroll, and the photo upload fails on 3G outside Lahore.

A few things I've learned the hard way running zivni with brands across eight countries:

Build the audit into the visit flow, not as a separate task. If a rep has to remember to "do the audit" after taking the order, half of them won't. The audit should be a step inside the visit — punch in, take order, do audit, punch out. One flow.

Pre-populate everything you can. The rep shouldn't be selecting their SKUs from a list of 400. The system should already know which SKUs that outlet stocks based on order history, and ask about those. Trust me, this single change can double completion rates.

Offline-first or you're dead. A lot of traditional trade outlets in Pakistan, Oman, even parts of the UK, have terrible connectivity. If your audit app can't work offline and sync later, your reps will skip audits in exactly the outlets where you need data most.

Give reps feedback they can use. Honestly, this is the part most brands forget. If a rep flags an OOS and nothing happens for three weeks, they stop flagging. Close the loop — show the rep what happened with their data, what got replenished, what got escalated. That's what keeps quality high.

And then on the management side — keep dashboards boring. A weekly view of audit completion by territory, OOS heatmap, share of shelf trend, planogram compliance score. That's 90% of what category managers actually look at. The other 10% is ad hoc questions you can pull when you need them.

A word on auditor bias

Look, here's something nobody likes to admit. When your own field reps do the audits, the numbers always look better than reality. Not because they're lying — well, sometimes — but because they're rushed, they round up, they assume.

The fix isn't surveillance. It's a mix. Have reps do the routine audits (because they're already there anyway, and it's cheap), but layer in a smaller sample of independent audits — third party, or your supervisors, or even just AI-verified photo audits — to calibrate. If rep-reported share of shelf says 34% and independent audit says 22%, you've got a coaching problem, not a data problem.

I got this wrong at first. Used to think rep-led audits were enough if you trained well. They're not. The calibration sample matters more than the training.

Start small. Pick three KPIs, twelve checklist items, one channel. Get it humming for 90 days. Then expand.

That's the order. Not the other way around.

© 2026 Zivni. All rights reserved.