Tools

Checklist for choosing an electronic shelf label vendor

Published November 9, 2026 Updated November 9, 2026 Written by: Editorial team

Published by Keonex. Not affiliated with etiquetaselectronicas.com. Keonex is a manufacturer and implementer of electronic shelf labels. How we work and why we say so: methodology and transparency.

This checklist scores an electronic shelf label (ESL) vendor across five blocks, nearly all built around one question: how much would it cost you to leave? Each block has five questions scored from 0 to 3. The result lets you compare offers on the same footing and decide what to negotiate before you sign.

What it is for, and what it is not

A label project is bought once and lived with for years. The labels hang on the shelves, the management software knows which price belongs on each one, and the contract decides who may touch what. If, with twenty thousand labels already installed, you discover that the data will not come out, that the API is a paid extra, or that another manufacturer’s labels cannot be mixed with yours, the vendor holds the negotiating position.

The checklist does not tell you which brand to buy. Nor does it replace the RFP template, the document you send to vendors. This does the opposite job: you receive the answers, check them against the contract and assign points. If you first need to understand why vendor dependence is a risk in this sector, the article on ESL lock-in covers it; here we only measure it.

How scoring works

Each question is worth 0 to 3 points, and the score comes from what the vendor gives you in writing, not from what was said in a meeting.

Points Meaning
0 Not offered, not documented, or the vendor refuses to put it in writing
1 Offered only at an uncapped extra cost, at the vendor’s discretion, or without public documentation
2 Offered and documented, with a limitation you have read and understand
3 Offered, documented, included in the price and written into the contract

There are 25 questions, so the maximum is 75. The blocks carry equal weight by default, but the weight column in the final table lets you raise the one that worries you most. A single-store retailer and a chain of forty stores do not fear lock-in to the same degree.

Block A. Data exit

Labels show a price, but the asset that matters lives on the platform: the assignment of each label to each item and shelf position, the template graphics, the change history and the update rules. If that cannot be extracted in a readable format, switching vendors means rebuilding it by hand.

  1. Full export. Can you download, at any time, with no ticket and no fee, the label-item-location assignment for all stores?
  2. Format. Is the export format open and documented (CSV, JSON, XML), or is it a dump that only the vendor’s own platform can import?
  3. Templates. Can you export the label display templates (layout, fields, rules) so that another system could reproduce them, even with some effort?
  4. History. Is the per-label price-change log by date, which you may need to justify a price in a dispute, exportable, and for how long is it retained?
  5. Written ownership. Does the contract state expressly that your store data is yours and that the vendor will not use it for other purposes without your authorisation?

It is tempting to give a 3 on question 1 because the platform has an export button. Before awarding the point, test it: ask for a real export from a demo environment and check that it opens and contains what you expect.

If the platform is a cloud service, switching vendors may also fall under the EU Data Act (Regulation (EU) 2023/2854). According to the legal analyses consulted, for data processing services within its scope the provider must support switching with a 30-day transition period, extendable to up to seven months where technically unfeasible, and switching charges are banned from 12 January 2027. Whether a label management platform falls within that definition is a legal question for your adviser, not this page; but it is a useful argument for demanding a clear exit clause. The primary text is on EUR-Lex, with commentary from Mishcon de Reya and Travers Smith.

Block B. Open APIs and integration

Your ERP or pricing system has to talk to the label platform. The integration is paid for once; what makes it expensive to change is when it is proprietary.

  1. Public documentation. Is the API documented in a place you can read without signing a contract, with examples and versions?
  2. Coverage. Can everything done in the interface (linking labels, changing prices, triggering updates, reading status) also be done through the API, or are some functions screen-only?
  3. Cost and limits. Is API access included in the price, and are the rate limits compatible with your peak updates (a bulk price change, a promotion)?
  4. Existing integration. Is there a ready-made integration with your ERP or POS, maintained by the vendor or a third party, and who fixes it when your ERP changes version?
  5. Connector ownership. If the vendor wrote the connector for you, do you hold rights to its code or specification so that someone else can maintain it?

Question 10 is the one most often skipped. A connector only the vendor can touch is dependence dressed up as an included service.

Block C. Label interoperability

This is the block where theory and reality are most easily confused. A public specification exists, the Bluetooth SIG Electronic Shelf Label (ESL) profile, which defines how a client can control and update labels over Bluetooth (version 1.0.1 on its official page, with conformance and test documents). But a standard existing does not mean the whole installed base uses it. US FCC equipment records hint at how far apart two brands can be: Hanshow’s STELLARP-154 label is certified in the 2402-2480 MHz band, while SoluM’s Newton EL043DWWRA sits at 902.4-927.6 MHz. Those are US variants, but the logic carries over: two labels that do not even transmit in the same band will not share a gateway. The public records do not say which protocol runs on top, and that is exactly the question to put to the vendor.

  1. Protocol. Does the vendor state in writing which radio protocol the labels use, and whether it conforms to the Bluetooth SIG ESL profile or is proprietary?
  2. Mixed brands. Can another manufacturer’s labels coexist in your store under the same platform, or only the vendor’s own?
  3. Third-party platform. Could the labels you buy be run by a software platform other than the manufacturer’s, if it came to that?
  4. Replacement stock. Does the vendor commit to selling labels compatible with the installed ones for a stated number of years, and what happens to models that are no longer manufactured?
  5. Infrastructure. Are the access points or radio gateways vendor-specific, or do they work with other labels? Who owns them?

If the vendor answers “proprietary, but stable and with guaranteed support”, score 2 on question 11 and let the score drop on the others: the block measures not the technology but how many exits it leaves you.

Block D. Contract exit terms

All of the above is worthless if the contract says otherwise. Read it with the next five questions in front of you, and if it is a lease or a per-label fee, read it twice; how that affects cost is covered in renting versus buying.

  1. Term and renewal. Does it renew automatically, what notice is needed to cancel, and at what price does it renew?
  2. Ownership at the end. Who keeps the labels, gateways and licences at the end: you, the vendor, or does it depend on paying a residual value?
  3. Exit clause. Is there a written procedure with deadlines for handing over your data and cooperating in a migration, and does it also apply if you terminate for the vendor’s breach?
  4. Exit price. Is the cost of exporting, uninstalling and removing already fixed in the contract, or left as “subject to quotation”?
  5. Unilateral changes. Can the vendor raise fees, withdraw features or change the platform during the term without your consent, and what termination right do you keep if it does?

A 0 here is not offset by a 3 in another block. It is the same logic as for data exit: what is not agreed before signing is rarely won afterwards.

Block E. Firmware and software dependence

An electronic label is a small computer with a screen and a battery. What it does depends on firmware, and whoever controls the updates controls how long your investment stays useful.

  1. Update policy. Is there a written commitment to update label and gateway firmware, including security fixes, for a stated number of years?
  2. End of life. What happens when a label generation loses support: does it keep working without updates, must it be replaced, and at whose cost?
  3. Licensing model. Are platform functions licensed per label, per store or per user, and can a licence lapse and leave the labels unmanaged?
  4. Offline operation. If the vendor’s cloud becomes unavailable, what happens to labels already loaded, and is there an on-premises option?
  5. Documented security. Does the vendor document how communication with the labels is encrypted, how keys are managed and how a vulnerability is reported?

The certification records cited earlier cover frequency bands, not encryption or key handling. That information rarely appears in any public register, which is why question 25 asks the vendor to put it in writing and commit to maintaining it.

Summary and reading the score

Record the points for each block (maximum 15) and multiply by the weight you choose. If you have no reason to do otherwise, leave 1 everywhere.

Block What it measures Points (0-15) Weight Total
A. Data exit Being able to take your information with you 1
B. API and integration Talking to your systems without a third party 1
C. Interoperability Room to change labels or platform 1
D. Exit terms What the contract says at the end 1
E. Firmware and software How long the installed base will stay alive 1

To read the result, do not stop at the sum:

  • Over 60 out of 75, with no 0: contained dependence risk. Still check that the answers are in the contract and not just in the sales email.
  • Between 40 and 60: acceptable dependence if offset by other factors, but every lost point is a clause to negotiate. Start with the questions scoring 0 or 1.
  • Under 40, or any block below 5: the vendor may be good at other things, but in this offer the exit is closed. Ask for changes in writing or compare with another proposal.

These bands are our own rule of thumb, not an industry standard. Adapt them to the size and risk of your project.

Using it in practice

  1. Send the 25 questions with the RFP and ask for numbered answers. When everyone answers the same thing, comparing is far easier.
  2. Score separately, with two people: someone from IT for blocks B, C and E, and someone from procurement or legal for A and D. Then reconcile the points.
  3. Verify one answer per block: a real export, a test API call, a contract excerpt. A vendor that lets you check is giving you information.
  4. Carry the results into the pilot. Open doubts can become criteria for the pilot project.
  5. Put the exit cost in the calculation. The total cost calculator measures the cost of buying and operating; this checklist’s score tells you what the freedom to leave is worth.

To see which manufacturers exist and what public data there is on them, see the manufacturer comparison.

Transparency

This site is published by Keonex, which manufactures and implements electronic shelf label systems. The checklist questions apply equally to Keonex and to any other vendor, and you should ask Keonex for the same written answers. This page does not score any specific vendor.

Frequently asked questions

What is the minimum score a vendor should reach?

There is no universal threshold. As a starting rule, treat any block scoring under a third of its possible points as a warning, and any "0" on data export or exit terms, which are the hardest things to fix once you have signed.

Are vendors with a proprietary radio protocol worse?

Not by definition. A proprietary protocol can be stable and well supported. What changes is the risk: if the labels only talk to one platform, changing platform means changing labels. The checklist scores that dependency rather than ruling it out.

Does the checklist work if the vendor rents me the labels?

Yes, and it deserves more rigour. In a rental the contract decides who keeps the labels, the data and the configurations at the end, which is exactly what the checklist examines.

Does this checklist replace an RFP?

No. The checklist is for evaluating answers and spotting gaps. The RFP is the document you send so that every vendor answers the same questions.

Sources

  1. Bluetooth SIG - Electronic Shelf Label Profile (version 1.0.1)
  2. Hanshow STELLARP-154 - FCC equipment record (2402-2480 MHz)
  3. SoluM Newton EL043DWWRA - FCC equipment record (902.4-927.6 MHz)
  4. Regulation (EU) 2023/2854 (Data Act) - EUR-Lex
  5. Mishcon de Reya - EU Data Act, switching rights for cloud services
  6. Travers Smith - The Data Act's changes to cloud services contracts

Work out the total cost of a project

A calculator with editable assumptions: labels, gateways, software, installation and a 60-month rental option. No vendor prices: you supply them.