Retail has instrumented everything except its own workforce
Ask a store manager which of their team members can safely operate the new self-checkout diagnostics, handle a chargeback dispute, or run a stocktake on the updated inventory system, and you'll usually get a confident answer. Ask them to prove it, and the confidence evaporates.
Most retailers still track skills through some mix of a manager's memory and a spreadsheet last touched in January, and the gap between what leaders believe their workforce can do and what it can verifiably do has become one of the most expensive blind spots in the sector.
The gap nobody can see
The scale of the problem is well documented. The World Economic Forum's Future of Jobs Report 2025 found that employers expect 39% of workers' core skills to change by 2030, and 63% of employers now name skill gaps as the single biggest barrier to business transformation. That puts skill gaps ahead of regulation and even ahead of access to capital.
Retail feels this more sharply than most industries. The technology stack inside an average store has changed faster than the job descriptions attached to it. Electronic shelf labels, RFID stock counting, mobile POS, computer vision loss prevention, in-store fulfilment for online orders: each roll-out adds a new set of competencies that someone on the shop floor needs to hold. Meanwhile, frontline turnover routinely exceeds 60% a year in some markets, which means the skills a retailer painstakingly built last quarter may have walked out the door before this one ends.
The result is a workforce picture that's permanently out of date. And you can't close a gap you can't see.
Start with a map, not a training budget
The instinctive response to a skills problem is to spend more on training. That's usually premature. Training budgets get spread evenly across teams because nobody knows precisely where the deficiencies sit, so high performers sit through refreshers they don't need while genuine risks, like a single certified forklift operator covering three shifts, go unnoticed until something breaks.
The better starting point is conducting a competency gap analysis: a structured comparison of the skills each role requires against the skills each person verifiably holds. Done properly, it turns a vague sense of "we need more digital skills" into a precise list of who needs what, and how urgently.
For retailers this matters in three areas, and they aren't equally urgent. Compliance is the least glamorous and the most important: food safety certifications, age-restricted sales training and equipment licences all expire on their own schedule, and an audit doesn't care that the spreadsheet was almost up to date. Deployment comes next, since scheduling someone onto a shift they aren't qualified to cover is an operational risk you only discover mid-shift. Retention is the slow burn. Employees who can see a mapped path from their current skills to the next role tend to stay longer, and in a sector where replacing one frontline worker costs thousands, that alone can pay for the exercise.
Where workforce technology fits in
Competency data becomes genuinely useful when it stops living in isolation and starts feeding the systems that run the store. Scheduling is the obvious example. When Erewhon, the Southern California organic grocer better known for its celebrity smoothie collabs than its back office systems, adopted Legion's AI driven workforce management platform, the pitch was demand forecasting and automated scheduling. But forecasting demand only solves half the problem. The other half is knowing which available employees actually hold the competencies each shift requires, and that answer has to come from somewhere more reliable than a manager's memory.
The same logic applies to task management, succession planning and store openings. A retailer expanding into a new region can pull up a competency matrix and immediately see whether it has enough certified people to staff the launch, or whether it's about to discover the shortfall two weeks before opening day. It appears simple once the data exists. Getting the data to exist is the hard part, which is exactly why the analysis has to come before the software decisions, not after.
The frontline is where this succeeds or fails
There's a trap here worth naming: competency programmes designed in head office often die on the shop floor. If logging a skill or completing an assessment requires a desktop computer and a spare half hour, frontline staff won't do it, and the data will rot within months.
The retailers getting this right are meeting employees where they already are, which is on their phones. When Flash Coffee rolled out YOOBIC's frontline employee experience app, gamified micro-learning sat alongside daily operational checklists in a single tool. That bundling is the point. Training that sits inside the daily workflow gets done. Move it to a separate portal and completion rates fall off a cliff.
Culture matters as much as tooling. Business leaders consistently report that efforts to upskill frontline staff only stick when employees can see the personal payoff, whether that's a clearer route to a supervisor role or recognition for skills they've already mastered. A gap analysis that reads as surveillance will breed resentment. One that reads as an investment in people tends to do the opposite, though it takes honest communication from managers to land that way.
Making it stick
The retailers that treat skills visibility as a one-off project will be back where they started within a year, because the workforce and the technology will both have moved on. The ones that treat it as infrastructure, refreshed continuously and wired into scheduling, compliance and career development, will make faster decisions with less risk.
The sector has spent a decade wiring sensors into footfall counters, shelf edges and fridge compressors. The workforce is the last major system still running on guesswork. Closing that visibility gap won't show up on a shelf edge or a sales dashboard, but it decides how well everything else in the store actually runs.
