The verdict in three sentences
To run 60 merchandisers across 4,500 stores in South Florida, a custom shelf audit app costs roughly USD 60,000 to 110,000 in 2026, for about 4 months of work. It must work offline (PWA), prove each visit with a geotagged photo and automatically compute share of shelf and out-of-stock rates per SKU. The return shows up in recovered sales: every out-of-stock caught and fixed in 48 hours instead of 10 days is worth far more than the app.
What the sales director needs every Monday
Spreadsheets texted on Friday night cannot compare banners or prove a visit happened. The app must produce reliable metrics calculated the same way everywhere, from a Publix in Coral Gables to a bodega in Hialeah.
| Metric | Definition | Typical FMCG target |
|---|---|---|
| Numeric distribution | % of stores where the SKU is present | 85% in grocery chains |
| Out-of-stock rate | % of visits where an expected SKU is missing | Below 5% |
| Share of shelf | Brand facings / total category facings | At or above market share |
| Planogram compliance | Match with the negotiated layout | 90% in key accounts |
| Shelf price vs suggested retail | Gap in USD and % | Below 5% |
| Visits done vs planned | Route coverage | 95% |
| Average visit time | Between geotagged check-in and check-out | 15 to 25 minutes |
Consolidated by banner (Publix, Winn-Dixie, Walmart, independents), these metrics become your leverage in annual line reviews.
Features and budget
| Module | Content | 2026 ballpark (USD) |
|---|---|---|
| Offline field PWA | Daily route, store cards, deferred sync | 14,000 to 24,000 |
| Audit forms | Presence, facings, price, expiry, POS material | 9,000 to 16,000 |
| Geotagged, timestamped photos | Compression, position check | 7,000 to 13,000 |
| Share of shelf and OOS engine | Rules per category | 7,000 to 14,000 |
| Route planning | Frequency by segment, absence reassignment | 6,000 to 12,000 |
| Dashboard per banner and territory | Maps, trends, OOS alerts | 10,000 to 18,000 |
| ERP or DMS integration and 4,500-store import | Product and customer master data | 7,000 to 13,000 |
| Total | 60,000 to 110,000 |
For comparison, off-the-shelf retail execution SaaS often runs USD 40 to 90 per user per month in 2026, so USD 29,000 to 65,000 a year for 60 users, with less control over metrics. Automatic image recognition (counting facings from photos) raises the bill and belongs in a second phase: assisted manual counting already yields usable data.
Offline and field reality
A merchandiser spends 6 to 8 hours a day inside stores where signal is often weak, especially in back rooms and big-box aisles. An app that waits for the server will be abandoned within two weeks.
| Constraint | Technical answer |
|---|---|
| Weak in-store signal | Full local storage, background sync |
| Data and battery use | Photos compressed to 150 to 250 KB, upload on Wi-Fi when possible |
| Mixed iOS and Android devices | Light PWA under 5 MB, no app store to manage |
| Visit fraud | Geofenced check-in within 100 m, timestamped photo |
| Team turnover | 3-screen interface, half-day training |
Build or buy: how to decide
A SaaS tool is the right call if you have fewer than 20 field users, standard metrics and no ERP integration needs. Custom development makes sense once you pass 50 users, need your own share of shelf rules per category, or want to resell reporting to brands under your own name. Over 3 years, 60 SaaS seats at USD 60 a month cost about USD 130,000, close to a custom build that you then own outright, with only hosting and support to pay afterwards.
Mini case study
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Carla, sales director of an FMCG distributor in Miami, tracks 120 SKUs in 4,500 stores. Today an out-of-stock is found on average 10 days after it starts. Across 600 key stores she estimates an 8% OOS rate and an average USD 12 of lost sales per SKU per day out of stock.
Calculation: 600 stores x 120 SKUs x 8% = 5,760 OOS situations. Detecting in 2 days instead of 10 recovers 8 days x USD 12, so USD 96 per situation. Even fixing only half, the gain approaches USD 276,000 in sales per annual OOS cycle. An USD 85,000 app pays back in under 4 months, before counting supervisor time no longer spent consolidating spreadsheets.
FAQ
Why a PWA rather than a native app?
A PWA installs from a link, updates without app store review and weighs under 5 MB. For 60 mixed devices, it is simpler to roll out and maintain.
How long does an in-store audit take?
With well-designed forms, 12 to 18 minutes for 30 SKUs, photos included. Routes usually plan 10 to 14 visits a day across Miami-Dade traffic.
Can we verify the merchandiser was really in the store?
Yes: geofenced check-in and check-out within 100 m plus timestamped photos make fake visits very hard. Anomalies go to the supervisor automatically.
Can data be shared with brands?
Yes, through a supplier space that only shows their own SKUs. Several distributors bill brands for this reporting.
What does maintenance cost?
Maintenance is a monthly support contract sized to your response-time needs; hosting and photo storage scale with volume, typically a few hundred USD a month.
Let's scope your project. We define your merchandising metrics, routes and store master data, then deliver an offline-first PWA in about 4 months. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
