# Electronic Shelf Labels and Black Friday: Mass Price Changes Without Errors

> How to change thousands of prices on Black Friday with electronic shelf labels: rehearsal, scheduling, verification, the prior price and a way back.

Source: https://etiquetaselectronicas.io/en/blog/electronic-shelf-labels-black-friday/
Published: 2026-11-09 | Updated: 2026-11-09

A mass price change on Black Friday goes well when the mistake is found on Tuesday, not on Friday at nine in the morning. That takes four things, in this order: fix the reference prices before the campaign starts, rehearse the full batch in one store, schedule it in waves, and check label by label.

The calendar and the checklist are below.

This text does not explain what electronic shelf labels are, nor does it set out the law in full. For the first, see the [guide to pricing with ESL](https://etiquetaselectronicas.io/en/pricing/); for the second, the page on the [30-day reference price](https://etiquetaselectronicas.io/en/regulation/was-now-pricing-reference-price-rules/). What follows is operations only: what to do, when, and how to check it.

A note on scope. The legal references here are EU rules, and the worked examples follow the Spanish transposition because that is the market we have sourced. If you trade in another member state, the principle is the same but the national wording and the exceptions can differ. The UK is outside this regime and is not covered.

## The calendar, with this year's dates

Black Friday 2026 falls on Friday 27 November and Cyber Monday on Monday 30 November. This article was published on 9 November. The count that matters does not start on the Friday but thirty days before each promotion begins:

| If the promotion starts on… | The 30-day window starts around… | Status on 9 November |
|---|---|---|
| Friday 20 November (an early "Black Week") | 21 October | Open for almost three weeks: whatever was applied since 21 October already counts |
| Friday 27 November | 28 October | Open for twelve days: almost half of it is already decided |
| Monday 30 November | 31 October | Open for nine days |

The practical consequence is uncomfortable. The prior price of every item you will advertise on the 27th is already decided by what you did between late October and today, including any Halloween or "early Christmas" promotion that lowered a price for a few days. What is left to control is what you apply between now and Friday: do not raise prices "to leave room", and do not launch small cuts that then set the reference lower than planned. The Directive defines the prior price as the lowest price applied in the 30 days before the reduction, and the Commission's guidance counts earlier offers in that minimum; in Spain the rule is in article 20.1 of Law 7/1996. The exact day count, that is, whether the first day of the window is the 28th or the 27th, is one of the points awaiting legal review.

The other end of the calendar has a date too. A reduction starting around 26 or 27 December, for example post-Christmas sales, will have the Black Friday price inside its window. If that was the lowest, it becomes the prior price. What you apply on 27 November limits what you can advertise a month later.

## Before: decide what is a sale and what is just a lower price

With this many labels and changes, the first mistake is mixing two different things in the same change file: items advertised as reduced and items that simply drop in price. The Commission's guidance treats as a price-reduction announcement not only "20% off" or "was 100, now 80" but also labels such as "Black Friday" or "deals" that give the impression of a reduction. When a label carries that claim, treat it as one.

For each item in the batch, prepare five fields, not three: new price, calculated prior price, percentage (if shown), campaign identifier and reason for the price. The reason matters because of a detail in the Spanish text: a price applied to cut food waste on products close to expiry does not count as the prior price, and the system can only exclude it if it knows that was the reason. Having all the fields means the calculation does not depend on someone doing it by hand at eleven at night.

If you advertise a percentage, it is calculated on that prior price. The Court of Justice made this clear in C-330/23 (26 September 2024): stating the 30-day minimum somewhere in the leaflet while calculating the percentage on the last price is not enough. A campaign file that arrives from marketing with the percentage already computed against the catalogue price is the most likely failure on this date. It is caught with a control column, file percentage against recalculated percentage, and every row where they differ gets reviewed.

## Rehearsal: one store, the whole batch

The rehearsal has to use the same file that will be launched, not a sample. A pilot store, or the smallest one, receives the complete batch a few days earlier at a time with no customers, and three different things are checked.

**That every price is the one it should be.** Compare a sample of the file with what the label shows. This catches format faults: shifted decimals, currency symbols, unit prices recalculated wrongly, or text that does not fit the label. It is what slips through when only the spreadsheet is validated.

**That the prior price is correct.** For a sample of items, the person in charge recalculates the 30-day minimum by hand from the history of applied prices and compares it with the one in the file. If the history does not let you do that in a few minutes, the problem is not Black Friday but the system, and it is better to know today.

**That the till charges the same.** A shelf showing 4.99 € and a till charging 5.49 € is a consumer incident, not a display fault. Run twenty or thirty items from the batch through the till, chosen among best sellers and among awkward formats (multipacks, price per kilo, "2 for 1" offers). The rules on which price prevails are in [shelf versus till](https://etiquetaselectronicas.io/en/regulation/shelf-price-different-from-till-price/).

The rehearsal also measures something nobody usually measures: how long the batch takes to finish in a real store. Manufacturers publish figures. A 2008 Pricer release, for example, claimed that a store of any size could complete a full update in roughly an hour, and that its platform exceeded 50,000 full updates per hour. These are vendor claims, more than fifteen years old, about another platform and with another definition of "update". They work as an order of magnitude, not as a guarantee. The time that counts is your store's, with your number of labels, your walls and your installation, and you get it by timing it.

## Scheduling: waves, not one button

The temptation is to launch everything at midnight on Thursday. That makes sense for e-commerce, where a price changes in an instant. In a physical store with labels there are two reasons to prefer another sequence.

The first is to separate the price from the claim. If the store wants to show "Black Friday" on the label, and the new graphic template can be loaded ahead of time without changing the price, do it in two steps: templates first (days before, no customer impact) and prices afterwards. That way only the minimum data travels on launch night.

The second is to launch in waves. A small group of stores first, a check, then the rest. If something fails, it fails in a few stores. For a single-store retailer the equivalent is launching one section first (electronics, say) and verifying it before the others.

What to schedule and what not to:

- The launch, at a time when the store is closed and there is room to resend. The time is set by what the rehearsal measured, plus a cushion.
- The return to the normal price, also scheduled with a date and time, so it does not depend on someone remembering on Monday morning. Offers left active by oversight end up as prices applied for weeks.
- Nothing that can be edited "by hand" on the label without going through the system. A change made outside the system leaves no history, and without history you cannot prove the prior price.

## Verification: what is checked and by whom

That the system says "sent" does not mean the label shows it. The check has layers, and each is done by a different profile.

1. **Confirmation report.** The label system normally produces a list of labels that have not confirmed the change. Review it before opening. Stragglers are resent, and those that still do not answer move to a physical list.
2. **Aisle walk.** A person with the list of the fifty highest-volume and most-promoted items checks the label in the aisle. It is a twenty or thirty minute walk.
3. **Till sample.** The same check as in the rehearsal, repeated with a different selection.
4. **Monitoring during the day.** Someone responsible for incidents on the shop floor: labels showing a prior price different from the offer, or customers pointing out a difference. Those calls are the fastest signal of a file error.

The price the customer sees is the price that must be charged. On paper or on a screen the obligation is the same, and authorities watch these dates. Spain's Ministry of Consumer Affairs announced in 2025 mass price monitoring to detect pre-Black Friday increases followed by apparent discounts, according to the note from the European Consumer Centre in Spain dated 19 November 2025. That note referred to large online companies; we have not found an equivalent announcement for 2026, nor a campaign aimed at physical stores. But the rule being supervised is the same.

## The way back: prepared before it is needed

A rollback is not the opposite of the launch. Changing the prices of a thousand items "back to how they were" can be as dangerous as the original change, because "how they were" has to be reconstructed. Prepare it in advance with three elements.

**A copy of the previous state.** Before launch, export the price state in force by item and store. This is the file you return to if you must revert.

**A criterion for what gets reverted.** Reverting everything is the last thing you want. Usually you revert a section, a store or a specific group of items. The rollback file must be generatable by those cuts and not only as a block.

**A record of what was applied.** Every price that was on a label, even for an hour, is kept with its start and end date and time. A wrong price that was visible, and lower than intended, raises a question we cannot answer with the information available: whether it counts as the prior price for later promotions. The rules speak of the lowest price "applied". The prudent course is not to depend on the answer, by avoiding the error with the validation above, and to consult a consumer-law lawyer if it happens.

## Checklist by date

| When | What | Who owns it |
|---|---|---|
| Now | Freeze price rises and small cuts on campaign items | Pricing manager |
| By 16 Nov | Campaign file with calculated prior price, recalculated percentage and reason | Pricing and marketing |
| Week of the 16th | Rehearse the full batch in one store; time it | Store operations |
| Week of the 16th | Load "Black Friday" graphic templates without touching prices | Whoever manages the labels |
| Wednesday 25 | Export the previous price state; set the rollback criterion | Pricing and systems |
| Thursday 26, after close | Launch in waves; confirmation report | Systems and operations |
| Friday 27, before opening | Aisle walk, till sample, resend stragglers | Store management |
| Friday 27 to Monday 30 | Incident monitoring; nobody edits by hand | Store management |
| Monday 30, evening | Return to normal price, scheduled, with the same verification | Systems and operations |
| Until about 27 December | Campaign prices are part of the window of any new reduction | Pricing |

The dates are a template: adjust them to your campaign and your store. If you have only one person for everything, the rehearsal and the till check are the last things to cut.

## If the foundation is missing: the pilot

Everything above assumes a system that keeps a history of applied prices and lets you simulate a batch, schedule by store group, confirm label by label and restore a previous state. That is a list of questions for any vendor, and they should be asked before signing, not in November. If you do not yet have labels and want to see how a mass change behaves with yours, a [pilot project](https://etiquetaselectronicas.io/en/implementation/pilot-project/) in one store is the way to measure it without risking the campaign.

The other question is whether the effort pays off. The saving per price change against paper depends on how many changes you make, and the [analysis of the cost of each change](https://etiquetaselectronicas.io/en/blog/electronic-shelf-labels-vs-paper-cost-per-price-change/) offers a method to measure it. For those who change prices during the day for other reasons, there are separate rules in [dynamic pricing](https://etiquetaselectronicas.io/en/regulation/dynamic-pricing-supermarkets/).

This article is not legal advice. The readings of the rule are under review and the doubtful points are flagged as such.

## Frequently asked questions

### When is Black Friday 2026?

Friday 27 November, with Cyber Monday on Monday 30 November. If a store's promotion starts that Friday, the 30 days that fix the prior price begin around 28 October; if it starts a week earlier, around 21 October.

### Do I have to show the prior price on the electronic shelf label during Black Friday?

In the EU, yes, for every item advertised as reduced. The prior price is the lowest price applied to identical goods in the preceding 30 days, earlier offers included. The full explanation is on the 30-day reference price page.

### How long does it take to update a whole store?

It depends on the manufacturer, the radio technology and the installation. Vendors publish figures of tens of thousands of updates per hour, or about an hour for a full store, but those are their own numbers, not independent measurements. What is useful is to time a test change in your own store and repeat it before the date.

### What if I get a price wrong and it is already on the labels?

Correct it as soon as possible and record what was applied, for how long and why. Whether that mistaken, lower price counts as the prior price for future promotions is a question for a consumer-law lawyer; this article does not settle it.

### Should every store be scheduled at the same time?

Not necessarily. Changing a pilot store or a small group first and verifying before releasing the rest lets you catch a configuration error while it affects a few labels instead of all of them.

## Sources

- [Commission Notice, Guidance on the interpretation and application of Article 6a of Directive 98/6/EC (2021/C 526/02)](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52021XC1229(06))
- [CJEU, Case C-330/23, Aldi Süd, judgment of 26 September 2024 (press release 152/24)](https://curia.europa.eu/jcms/upload/docs/application/pdf/2024-09/cp240152en.pdf)
- [Ley 7/1996, de 15 de enero, de Ordenación del Comercio Minorista, consolidated text, art. 20 (BOE-A-1996-1072), in Spanish](https://www.boe.es/buscar/act.php?id=BOE-A-1996-1072)
- [CEC-España, Recomendaciones para consumir de forma responsable durante el Black Friday (information note, 19.11.2025), in Spanish](https://portal-cec.consumo.gob.es/sites/default/files/documentos/NI_Black_Friday_21_11_2025.pdf)
- [Pricer, ESL platform breakthrough quadruples communication speed with existing infrastructure (11.02.2008)](https://www.pricer.com/press-release/pricer-esl-platform-breakthrough-quadruples-communication-speed-with-existing-infrastructure)