Electronic shelf label savings calculator over 60 months

This page works out, over 60 months, whether an electronic shelf label project pays for itself. It adds up the cost (labels, gateways, software, installation, integration, replacements) and sets it against the labour saved on price changes and the pricing errors avoided. The starting values are assumptions, not anyone’s price list.

What this page does and does not do

Project cost is broken down line by line in the pricing and total cost guide, and the lease-or-buy arithmetic is in lease vs buy. We do not repeat any of that here. What we add is the other half of the sum: what you stop spending, which almost no quote works out for you.

The total cost calculator has a form for the cost fields and runs in your browser. This page is its extended manual: it explains the savings model, works through one case with every number visible, and shows how far the result moves when an assumption turns out wrong. You can rebuild all of it in a spreadsheet from the formulas below.

The model in five lines

Everything comes down to two annual amounts and one up-front investment.

1. Upfront cost = labels × price per label + gateways × price per gateway + labels × installation cost per label + data integration.
2. Annual running cost = software fee × 12 + labels × price per label × annual replacement rate.
3. Annual time saving = price changes per week × 52 × (minutes per paper change − minutes per label change) ÷ 60 × cost of an hour.
4. Annual error saving = pricing incidents per week × 52 × cost per incident × share the label removes.
5. Result at 60 months = 5 × (time saving + error saving − annual running cost) − upfront cost.

Payback is the upfront cost divided by the net annual saving. None of these lines is discounted: the model is deliberately simple so it can be audited by hand. If your finance team works with net present value, apply its discount rate to the annual cash flow.

The assumptions and where they come from

Every default shows its origin. Where it says “use your own”, we found no open source to support the figure and will not make one up.

Figures are shown in pounds for readability. The public label and gateway list prices we cite in the pricing guide are in euros, so treat the unit prices below as the same order of magnitude, not as a currency conversion, and replace them with your quote.

Assumption	Example value	Origin
Labels in the store	5,000	Illustrative. Count by product lines and shelf faces.
Price per label	£10	Range of dated public distributor lists in the pricing guide, read across at face value. Your quote rules.
Gateways	6 at £300	Illustrative, in line with a catalogue model cited in that guide. Set by the coverage survey.
Installation	£0.50 per label	Illustrative. Ask your installer for hours and hourly rate.
Software	£100 a month	Illustrative. No open tariff exists that we could cite.
Data integration	£5,000 one-off	Illustrative. This is the line that varies most.
Annual replacement	1.5%	Illustrative. Ask for the committed failure rate.
Price changes per week	400	Use your own: it comes from your pricing system’s history.
Minutes per paper change	2.0	Use your own. Time it in a trial: print, locate, remove, place and check.
Minutes per label change	0.2	Use your own. Includes the occasional spot check, not the update itself, which is automatic.
Cost of an hour	£18	Illustrative. Use the fully loaded employer cost (wages, employer National Insurance, pension, holiday), not the pay rate. The statutory floor is the National Living Wage, £12.71 an hour from April 2026 for workers aged 21 and over.
Pricing incidents per week	20	Use your own. Take it from complaint counts and shelf-to-till differences.
Cost per incident	£8	Illustrative. Till time, refunding the difference, lost sale.
Incidents the label removes	70%	Illustrative. It does not remove those caused by wrong master data.

A note on errors. In 1998 the US FTC measured 3.35% of the items it checked as having a different price on the shelf and at the till, and reported that on non-sale items the errors came largely from incorrect shelf and item prices. It is an old study, from another country and another kind of retail. We cite it as a reminder that the failure exists and has been measured, not so that you can carry that percentage over to your store. Measure your own with an audit of a hundred lines. What follows when a customer spots it at the till is covered in the page on shelf price different from till price.

Worked example

A store using the values in the table above.

Item	Calculation	Amount
Labels	5,000 × £10	£50,000
Gateways	6 × £300	£1,800
Installation	5,000 × £0.50	£2,500
Integration	one-off	£5,000
Upfront cost		£59,300
Software	£100 × 12	£1,200 a year
Replacement	5,000 × £10 × 1.5%	£750 a year
Running cost		£1,950 a year
Time saving	400 × 52 × 1.8 min ÷ 60 × £18	£11,232 a year
Error saving	20 × 52 × £8 × 70%	£5,824 a year
Gross benefit		£17,056 a year
Net benefit	17,056 − 1,950	£15,106 a year

Over 60 months: 5 × 15,106 − 59,300 = £16,230 positive. Payback is 59,300 ÷ 15,106 = 3.9 years, about 47 months. A project that pays back in year four of five leaves little margin: any slip spoils it. That is exactly what happens in the next section.

Sensitivity: what moves the result

We change one assumption at a time and leave the rest as in the example.

Change	Result at 60 months
Base example	+£16,230
Minutes on paper: 2.0 → 1.0 (saving 0.8 min per change)	−£14,970
Changes per week: 400 → 200	−£11,850
Cost of an hour: £18 → £14	+£3,750
Price per label: £10 → £7 (replacement falls to £525 a year)	+£32,355

Read the table with one idea in mind: time assumptions outweigh hardware assumptions unless you secure a very different label price. If you are unsure how long your team takes to change a price today, spend a morning measuring it before asking for quotes. It is the cheapest figure to get and the one that skews the result most.

Changes per week depend on the kind of store. A shop with frequent promotions, seasonal lines or prices that move daily will have many; a hardware store with a stable price list, few. The rules on promotions add work to every paper price change. Britain has no statutory 30-day rule, but a was/now price still has to be a genuine one: the CTSI Business Companion guidance and the Digital Markets, Competition and Consumers Act 2024 treat a misleading reference price as an unfair practice. Where you also trade in the EU, the 30-day prior-price rule from the Omnibus Directive applies, explained in the 30-day reference price guide.

The limits of the model

A model like this misleads if you use it without knowing what it leaves out.

• Hours are not pounds. Time savings turn into money only if you reduce contracted hours or free someone to sell or restock. Otherwise it is capacity, not cash. Many projects are justified with hours nobody cuts afterwards.
• No discounting and no inflation. A pound saved in month 50 is worth less than one spent today. With high interest rates the real payback stretches.
• Errors do not disappear entirely. An electronic shelf label shows exactly the data it receives. If the master price is wrong, it will show it wrong, larger, and across the whole store at once.
• Hard-to-measure benefits are missing: sales from well-applied dynamic pricing, fewer gaps from pulling out-of-stock items, a better image. We leave them out because we cannot support them with a source. If you add them, document where each figure comes from.
• Hard-to-measure costs are missing: training, shelf rails, network, battery disposal. They are listed in the pricing guide.
• One store, not a chain. In a chain, integration is paid once and spread across stores, and the calculation changes shape.
• Obsolescence. Five years is the model’s horizon, not a guarantee of label or battery life.

How to use the result

Do the sum with your own data first and check it stays positive in a pessimistic scenario, not just the central one. If the project only works on the optimistic assumption, it is not a good project, it is a bet. Then ask each vendor for the same breakdown for a pilot store and recalculate. The implementation guide explains how to prepare the data and the pilot so the savings can be checked afterwards, not just promised.

If an offer arrives with a guaranteed saving and no account of how many changes, how many minutes and at what cost, it is a sales figure, not a calculation.