Summary. Retailers evaluate store devices on specification and cost, and then discover that the number which actually matters is how long it takes an associate to become competent with one. On current evidence that is measured in weeks, not hours, and in a workforce that turns over annually it is a recurring cost rather than a one-off. This paper puts a figure on that ramp, answers the obvious objection to treating it as a procurement question, and sets out what to measure before the next fleet decision is signed.

The number nobody puts in the business case

Scandit surveyed 2,000 store associates across seven countries (the US, UK, Germany, France, Spain, Italy and Japan) spanning nine retail sub-sectors, with fieldwork run from November 2023 to January 2024. The sample deliberately included gig workers, full-time employees and fixed-term contractors, and excluded checkout staff.¹

The finding that should change how devices are procured concerns time to competence:

Require two weeks to use their devices effectively	34 percent
Require up to four weeks	30 percent
Gig workers	Reach competence roughly one week faster

Nearly two thirds of associates need between two and four weeks to work effectively with the tools they are handed. In a business where a large share of hourly staff turns over within a year, that ramp is paid again and again, and it is almost never costed in the device decision.

It is worth being precise about what the figure describes. It is not the time to complete onboarding, and it is not the time until someone can log in. It is the associate’s own account of when they could work effectively with the device in front of a customer. That is the only definition that matters commercially, because everything before it is a period in which the store is paying full wages for partial output.

What two to four weeks actually costs you

The reason this number stays out of business cases is that nobody is asked to produce it. The hardware quote is a line item with a supplier attached. The ramp is diffuse, sits in store labor, and belongs to no one. It is also, in most fleets, the larger figure.

The arithmetic is four steps and you already hold every input.

1. Annual new starters across the fleet. Not headcount. The number of people who joined in the last twelve months, which in hourly retail roles is frequently a large fraction of the total.
2. Days to competence. Use your own figure if you have one. If you do not, the two to four week range above is a defensible placeholder, and the fact that you are using someone else’s number is itself the finding.
3. The productivity shortfall during the ramp. Not one hundred percent. A new associate is producing something from day one. The honest question is what share of a competent associate’s output they deliver in week one versus week four.
4. Loaded hourly cost. Wages plus the supervisor time that a struggling new starter absorbs, which is the part usually left out and is frequently the more expensive half.

Multiply those four and you have the annual cost of your current ramp. Then run the same calculation with the ramp one week shorter. The difference is what a faster device is worth per year, and in most fleets of any size it is a larger number than the price gap between the two devices being compared.

That is the calculation to bring to the procurement conversation, because it reframes the decision. The question stops being which device costs less to buy and becomes which device costs less to learn.

“That is a training problem, not a device problem”

This is the fair objection and it deserves a direct answer rather than being waved past. If associates need a month to become competent, perhaps the onboarding is thin and the device is fine.

Three things in the evidence argue otherwise.

Associates name intuitiveness as a device attribute. When asked what they value, 52 percent said the device being intuitive, ranking it alongside functional capabilities rather than alongside training.¹ The people doing the learning are describing a property of the tool.

Gig workers ramp faster on the same instruction. They typically receive less onboarding, not more, and still reach competence about a week sooner.¹ If curriculum were the binding constraint, that result would run the other way.

Training time is not free either. Every hour added to onboarding is an hour bought at the same wage, competing with selling hours, and delivered by a store manager who has other obligations. “Train more” is not a way of avoiding the cost. It is a way of moving it into a budget where nobody will attribute it to the device.

The strongest version of the counter-argument survives in a narrower form, and it is correct: training and tool design are not substitutes. A well-designed device with no briefing still fails, and the fastest ramps come from pairing a tool that behaves predictably with content that reaches the associate on the same device they are learning to use.

What associates actually value in a device

The same research asked what capabilities matter, and the answers are notably unglamorous.¹

• The ability to multi-task, 70 percent. A device that forces an associate to close one thing to do another loses to one that does not.
• Access to product information, 67 percent. The single most requested capability is the one directly served by training content.
• Device intuitiveness, 52 percent. Which is the same statement as the ramp figure above, expressed as a preference.

Over two thirds of associates rate their device as important or very important to doing their job. This is not a workforce resisting technology. It is a workforce asking for tools that work.

Notice what is absent from that list. No one is asking for more features. The three things named are speed of access, breadth of answer, and the absence of friction, which are the characteristics of a tool someone uses fifty times a shift rather than one they are shown in a demonstration.

The gap they report

40 percent believe their employer does not invest in their technology needs, and 20 percent feel it is treated as a non-priority.¹

That perception has a retention dimension. When associates were asked what drives loyalty, modern and easy-to-use technology was named by 41 percent, behind work-life balance at 56 percent and competitive salary at 54 percent, but ahead of much of what usually occupies engagement programs.¹ Tools are a working condition.

This closes a loop worth naming. A slow ramp lengthens the period in which someone feels incompetent in front of customers, and feeling incompetent in front of customers is a common reason for leaving early. Leaving early raises the new-starter count, which is the first input in the arithmetic above. The device decision and the turnover figure are connected, and they are usually reviewed by different people in different quarters.

Why the ramp is mostly software

Two identical handhelds can produce very different ramp figures, because what an associate is actually learning is a set of applications, conventions and logins rather than a piece of hardware.

The multiplier is the number of separate tools someone must hold in their head. Each additional application has its own navigation, its own vocabulary for the same object, and its own failure mode when the network is poor. Four applications is not four times one application to learn. It is worse, because the associate also has to remember which one answers which question, and that decision happens while a customer waits.

Consolidation is therefore a training decision that is usually made on an IT agenda. The device that ramps fastest is rarely the one with the best specification. It is the one where the answer to the most common question is reachable without choosing an application first.

What this means for the decision

1. Cost the ramp, not just the hardware. Run the four-step calculation above before the specification comparison, not after it.
2. Test with new hires, not with your project team. The people evaluating in a pilot are motivated and supported. The person who matters is a part-time associate in week one.
3. Judge the app, not the device. Ramp time is mostly a software property. The same hardware running a well-designed app is a different product from the same hardware running four unrelated ones.
4. Count the number of apps an associate must learn. Every additional tool has its own ramp, its own login and its own conventions.
5. Ask what happens when the device is shared. Most store devices are, and shared-device patterns break assumptions built for personal ones.
6. Write the ramp target into the contract. If a supplier believes their tool is intuitive, they should be willing to be measured on when your new starters say they became competent.

The BYOD question, briefly

The gig-worker finding is suggestive: they reach competence about a week faster, and one plausible explanation is familiarity with their own hardware.¹ That is an argument for interfaces that behave like the phone in someone’s pocket, whether or not it is literally their phone.

It is not, on this evidence, a settled argument for a bring-your-own-device policy. The research measures time to competence, not the security, wage-and-hour and support implications of personal devices. Those sit in their own decision.

What to measure

1. Self-reported days to competence, asked of every new starter at day thirty. One question, asked consistently, gives you the baseline that no vendor can supply.
2. The spread between your fastest and slowest stores. Same device, same onboarding, different managers. A wide spread is a coaching finding rather than a hardware one.
3. Applications opened per shift, per role. If the number is above three for a sales associate, consolidation is where the ramp is hiding.
4. Support contacts in the first thirty days. The questions new starters ask are the specification for the training you have not written yet.
5. Ninety-day retention against days to competence, by store. If the stores with the longest ramps also lose the most people, the device conversation has just become a labor conversation.

Where to start

Ask your last twenty new hires one question: how long before you felt confident using the store systems? You will get a better answer than any vendor’s demo provides, and you will be able to compare it against the two-to-four week benchmark above.

Adoption is the other half of this problem, and it is decided before deployment rather than after. The calculator will model what a shorter ramp is worth across your own fleet.

References

1. Scandit, Frontline Retail Revealed: Motivations, Technology Attitudes and Insights, published March 2024; fieldwork November 2023 to January 2024. 2,000 store associates across the US, UK, Germany, France, Spain, Italy and Japan, spanning nine retail sub-sectors, including gig, full-time and fixed-term workers and excluding checkout staff. scandit.com