Writing · Checklist
Electronic shelf label pilot checklist: what to prove before rollout
An electronic shelf label pilot should prove four things before you sign for every store: the shelf price always matches the POS, the system works in your hardest departments, a bad promotion can be reversed quickly, and a normal store can run it without the project team. Label hardware is the easy part. The twelve checks below cover what tends to fail once labels meet real shelves, real pricing rules and real staff.
Published 2026-11-22. Written for retail operations, pricing and store IT teams. We build Wi-Fi color e-paper signs, not price-rail labels, so we have not run an ESL pilot ourselves. This checklist is assembled from the published guidance linked below, and several of those sources are ESL vendors, which we flag where it matters.
The 12-point ESL pilot checklist
Use the right-hand column as your evidence. If you cannot point to it at the end of the pilot, the check has not passed. Set your own numeric thresholds before go-live: the sources we read do not publish a benchmark that would fit every retailer.
| Check | What to prove | Evidence |
|---|---|---|
| Price authority | One named system publishes the shelf price | Written decision, signed by pricing and IT |
| Master data | Unit prices, units of measure and deposits audited | Exception list closed for pilot categories |
| Label link | Every label maps to the right SKU and position | Sample audit: label, shelf price and POS price agree |
| Coverage | Radio tested with shelves fully stocked | Update results for freezer aisles, corners and back walls |
| Hardware fit | Size, enclosure, mount and temperature by department | Spec sheet matched to each department |
| Templates | Required fields and fallbacks per category | No blank fields when test data is missing |
| Conflicts | Promo start, price rule and clearance collisions | Scripted cases run, with expected and actual price |
| Rollback | A wrong promo can be reversed everywhere | Timed drill, from detection to corrected shelf |
| Exceptions | Dead battery, broken label, wrong assignment | Named owner and a store procedure for each |
| Metrics | Status per label, reviewed by store and category | Dashboard or report in use before go-live |
| Control | One store runs without extra project support | Same metrics, compared with the pilot stores |
| Rollout gate | Acceptance criteria set for each wave | Go or no-go recorded against the criteria |
ESL pilot: A limited deployment of electronic shelf labels in real stores, run to test the whole chain from pricing system to shelf, with acceptance criteria agreed in advance. It is a decision tool, not a soft launch.
Before the first label goes up
Name one source of truth for price
Decide which system publishes the shelf price before you compare hardware. Maximilian Kohl’s decision-maker guide in EuroShop.mag is firm on this: one system is the source of truth, and nothing bypasses it. It also recommends testing POS-to-shelf synchronization directly, comparing real-time and batch updates, and setting up variance reports with a clear escalation path.
Audit master data early
Paper tags hide messy data because someone fixes it by hand at print time. Labels publish it straight to the shelf. Inconsistent unit prices, units of measure and deposit values show up immediately, so audit them for the pilot categories first. Agree how unit prices are calculated across g, kg, ml and l, including rounding.
Record a baseline
A pilot cannot show savings against a number nobody measured. Before installation, record how long price changes take, how often shelf and POS disagree, and how many promo changeovers miss their start time. A planning guide from Racklify names price changes automated, labor hours reduced, price error reduction and sales impact as the pilot KPIs. Each needs a “before” figure. Our ESL break-even worksheet shows how those inputs feed the cost case.
Choosing pilot stores and departments
Pick stores for difficulty, not convenience. A pilot covering only dry ambient aisles misses the problems that appear in freezers and meat cases, as MinewTag’s rollout guide warns (MinewTag sells ESLs, so read it as vendor guidance). Include:
- A freezer or chilled aisle: cold shortens battery life, and refrigeration and metal affect the radio.
- A fresh department: weighed produce, markdowns and short bakery windows need more data than a small label may fit.
- A high-promotion category, where a timing mistake reaches every facing at once.
- A shelf section that gets reset often, so you test how labels are moved, reassigned and audited.
- Back walls, corners and seasonal areas, tested when fully stocked rather than on an empty floor plan.
Scale varies widely among published examples. Asda’s 2024 trial put 3,000 Vusion labels into one high-footfall Express store for 12 weeks. Tesco evaluated ESLs with VusionGroup and Hanshow in a 2025 proof of concept, and its Hanshow rollout starts in four stores on the way to about 3,000. A single complex store can teach you a lot; it cannot tell you how the fiftieth store will behave.
Network design is the other early decision. If you have not yet chosen between dedicated base stations and radios in your access points, read ESL base station vs Wi-Fi before you scope the pilot.
What to test during the pilot
Script the price conflicts
EuroShop.mag’s example is a promotion starting at 8:00 a.m. while a new pricing rule takes effect. Write the collisions down and run them on purpose: promo on top of a base price change, loyalty price against store-specific price, multi-buy ending at midnight, clearance on an item already on promotion. For each, note the price you expect at the shelf and at the till, then check both.
Drill the rollback
Paper mistakes stay in one aisle. An ESL mistake can reach every shelf position at once. Push a deliberately wrong test price to a contained set of labels, then time how long it takes to detect and reverse it. Include who has the authority to stop a change.
Run the boring failures
Assign an owner for dead batteries, cracked labels, wrong assignments and labels that fall off during a reset. The store team needs to know what to do at 7 a.m. when a label is blank and the vendor’s support line is closed. Decide what the fallback is at the shelf, and keep a small stock of spare labels on site.
Test readability at real distance
Check font size, contrast and icons from where shoppers actually stand, including low and high shelves. Keep price, unit price and promo terms clear. If a fresh item needs more information than the label can hold, move it to a nearby sign or a QR code rather than shrinking the type. That boundary between label and sign is covered in ESL vs digital signage.
How to decide after the pilot
EuroShop.mag cautions that pilot stores get far more attention than a normal store will, so add a store that gets no extra support and compare it with the others. Review results by store and category, because a healthy network average can hide one department that never updates reliably.
- Go: every check has evidence, and the control store performs close to the pilot stores.
- Go, narrower: the system works in ambient aisles but not cold ones. Roll out where it passed and keep paper or another label type elsewhere.
- Fix and repeat: failures trace to master data or process. Fix them and rerun the same scripts before adding stores.
- Stop: price accuracy cannot be proven, or the savings do not survive once real support costs are counted.
Plan the rollout in waves with acceptance criteria for each, and feed what each wave teaches back into templates and procedures. Racklify’s guide allows 6 to 12 weeks for rollout planning and months to a year for the rollout itself. For budget context, see our e-paper signage cost breakdown, and for brand-owned displays at the shelf, the retail shelves use case.
FAQ
How long should an electronic shelf label pilot run?
Long enough to cover a full promotion cycle, a price change from head office and your hardest departments. One published planning guide suggests 4 to 12 weeks of pilot deployment followed by 3 to 6 months of evaluation. Asda's 2024 trial at one Manchester Express store was planned for 12 weeks. Treat these as reference points, not a benchmark.
How many stores should be in an ESL pilot?
There is no standard number. Asda started with one store and 3,000 labels; Tesco's rollout project starts in four stores, one Express and three supermarkets. What matters more is that the stores cover your hard conditions, and that at least one store runs without extra project support so you can see how the system behaves on a normal week.
What KPIs should an ESL pilot measure?
Operational: update success rate, labels offline, price variances between shelf and POS, and battery status, reviewed by store and category. Business: price changes automated, labor hours saved, price errors reduced and any sales effect. Record the baseline before the labels go up, or the comparison will not hold.
What usually goes wrong in an ESL pilot?
Published guidance points to the same few causes: bad product-to-label links, inconsistent master data such as unit prices, weak radio coverage once shelves are fully stocked, cold departments that shorten battery life, and promotions that collide with price changes. A pilot limited to dry ambient aisles will not surface most of them.
Should the ESL pilot include refrigerated and freezer aisles?
Yes, if you plan to deploy there. Cold shortens battery life, metal and refrigeration affect radio coverage, and washdown areas need the right enclosure and mounting. Check the label's rated temperature range for each department and test in the actual case rather than assuming the dry-aisle result carries over.