Retail has access to more real-time demand data than almost any other frontline industry. Footfall counters, point-of-sale feeds, loyalty data, weather forecasts, local events – all of it can tell retailers more about when customers will arrive and what they are likely to buy.
Yet the same problem persists: too many people on the shop floor when demand is quiet, and too few when it peaks.
That matters even more as retail heads towards its most important trading period. Black Friday and Christmas are now only a handful of planning cycles away, while this year’s back-to-school season is already providing a smaller-scale preview of the same challenge.
The question is no longer whether retailers have enough data to anticipate demand. It’s why all that data still so often fails to produce the right schedule.
The stakes are significant. The National Retail Federation expects overall US retail sales to grow 4.4% in 2026, while holiday e-commerce spending alone is forecast to rise 7–9%, reaching as much as $310.7 billion.
At that scale, even a short period of poor staffing carries a commercial cost: missed sales when stores are stretched, wasted labour spend when they are quiet, or – as is often the case – both on the same day.
And that is the paradox. Retail should be one of the industries best equipped to match labour to demand. So why does it keep getting the balance wrong?
Research suggests this is not simply the result of the occasional bad rota.
A landmark study published in Production and Operations Management, based on hourly traffic, sales, and labour data from 41 stores in a large retail chain, found that every store was systematically understaffed during its own three-hour daily peak. At other points in the day, those same stores carried more labour than their traffic justified (Mani, Kesavan & Swaminathan).
A 2025 review of retail labour scheduling from Harvard Business School and the University of Texas at Austin revisits the same problem, linking it to the way forecasting and scheduling decisions are commonly made across retail organisations (Raman & Kwon, 2025).
The commercial consequences run in both directions. Analysis summarised by retail analytics firm Retail Sensing suggests that eliminating understaffing could increase sales by 7% and profitability by 5.7%, while overstaffing can reduce profitability by around 2%. Research cited in the same analysis found that a 1% increase in staffing levels could lift conversion by roughly 0.5 percentage points.
For an industry where labour typically represents 10–15% of revenue, repeatedly putting those hours in the wrong place is far from a marginal problem. It is one of the biggest controllable levers available to retail operators.
The tension is visible at market level too. Dutch retail sales grew 3.6% in 2025, yet Statistics Netherlands’ staffing sentiment indicator for the sector turned negative in the final quarter and was expected to weaken further into 2026 (CBS).
Demand can be growing while confidence in staffing falls. The problem is not always having too few labour hours overall. It is having the right labour, in the right place, at the right time.
Retail does not need to wait until Black Friday to see this dynamic play out.
In the US, the National Retail Federation expects K–12 back-to-school spending to reach a record $43.3 billion in 2026, up from $39.4 billion in 2025 and above the previous high of $41.5 billion in 2023. Combined back-to-school and back-to-college spending is expected to exceed $140 billion.
The shape of that demand is changing too. By early July, 62% of US shoppers had already started buying, stretching what might once have looked like a concentrated seasonal spike across a longer and less uniform trading window.
In the UK, back-to-school remains the second-largest retail event of the year after Christmas, worth an estimated £2.3 billion (SPS Commerce). According to seasonal spending analysis cited by Sutton Commerce, 62% of parents began shopping in late July and 85% are expected to have finished by the second week of September, with the busiest trading week falling in the final full week of August.
This is a demand peak retailers know is coming. But with UK footfall running below previous-year levels for much of 2026, simply adding more labour is not the answer. Stores need greater precision in when and where those hours are deployed (BRC-Sensormatic footfall data).
Similar pressures are playing out across the Nordics, DACH, and the Netherlands in categories such as clothing, stationery, electronics, and footwear.
Back-to-school may be smaller than the holiday season, but that is exactly what makes it useful as a warning. If retailers struggle to align staffing with a familiar and relatively predictable late-summer peak, Black Friday and Christmas will expose the same weaknesses on a much larger scale.
If retail’s problem were simply a lack of data, more sensors, feeds, and dashboards would have solved it years ago.
They haven’t.
The real problem is what happens between recognising demand and turning that insight into a schedule.
Legacy scheduling often begins with historical averages: last year’s sales for the same week, a weekly labour budget, or expected demand distributed relatively evenly across opening hours.
But averages smooth away precisely the variation that matters most.
A peak and a lull can cancel each other out on a weekly forecast while requiring completely different staffing levels on the shop floor. The result is a schedule that looks sensible in aggregate but is wrong at the moments that matter.
For predictable events such as back-to-school, Black Friday, or Christmas, that becomes particularly costly. The peak is known in advance, yet an average-led schedule can still treat it too much like an ordinary week.
Point-of-sale systems, footfall counters, loyalty platforms, workforce systems, weather data, and local event information are often accurate individually but disconnected operationally.
A change in customer traffic can be visible in one system without having any meaningful route into the next schedule. Turning that signal into a staffing decision then relies on someone spotting it, interpreting it, and manually adjusting the plan – often in a spreadsheet and under time pressure.
The issue is not visibility. It is connectivity.
Labour is typically controlled as a cost: wage percentage, scheduled hours, or adherence to a staffing budget. Sales, conversion, and customer experience may sit elsewhere in the organisation against a different set of targets.
That creates an obvious blind spot.
A store can technically hit its labour budget while losing sales because too few people are working during the rush. Equally, it can protect service levels while spending too much on quiet periods.
Both teams can optimise their own number without anyone optimising the relationship between the two.
And in markets where labour itself is becoming harder to secure, there is even less room for error.
German retailers were unable to fill around 122,000 vacancies in 2024, according to Handelsverband Deutschland, with the sector facing an increasingly significant demographic skills gap.
Across the Nordics, retailers face a similar imperative to generate more from the workforce they already have. Trade is Sweden’s largest private-sector industry, employing more than one in ten workers (Svensk Handel). In Finland, retail sales are expected to continue growing through 2026 even as employment declines: around 8,000 retail jobs disappeared in 2025, with another 12,000–13,000 forecast to go by 2027 (Kaupan liitto). Norway’s retail body Virke is also forecasting continued growth in 2026 as household purchasing power improves (Virke).
When retailers cannot simply add headcount, the value of every available labour hour rises.
The challenge also looks different by category. In grocery, where margins are tight and demand can shift quickly with weather, promotions, and local events, poor scheduling can mean queues building within minutes and underused tills shortly afterwards.
In fashion and apparel, the risk is often concentrated into a handful of highly valuable trading periods – back-to-school, Black Friday, Christmas, or end-of-season sales. Get staffing wrong on those days and there may be little opportunity to recover the lost revenue later.
Retail does not need another layer of information. It needs the data it already has to influence staffing decisions at the speed and level of detail at which demand actually changes.
That means moving beyond weekly averages and forecasting hour by hour – or, where appropriate, in even smaller intervals.
It means bringing signals such as sales, footfall, weather, promotions, and events into the same demand forecast instead of reviewing them across separate reports.
And it means giving finance and operations a shared view of what good staffing looks like, so the conversation moves beyond a trade-off between labour cost and customer service towards a more useful question: how much labour does this period of demand actually require?
AI is increasingly making that possible.
McKinsey’s recent research into AI adoption in European retail found measurable gains where AI-driven forecasting and planning have been properly connected to operational decision-making, including capacity gains of around 40% in functions moving from manual, average-based forecasting towards AI-driven models.
The important point is not that AI magically makes retail demand predictable.
It is that retailers can now process far more of the signals they already collect, at far greater granularity, and translate them into operational decisions quickly enough to matter.
That is what starts to resolve the scheduling paradox.
Back-to-school is the preview. Black Friday and Christmas are the real test.
Neither is an unexpected demand shock. Retailers know they are coming, they have years of historical data to draw from, and they increasingly have real-time signals showing how demand is changing.
The competitive difference will be whether that intelligence remains trapped in dashboards and spreadsheets – or whether it actually reaches the schedule before the next customer walks through the door.
Download Quinyx’s retail whitepaper to see how AI-powered demand forecasting can help you align staffing more closely with demand ahead of Black Friday and the Christmas trading period.