Two ways of deciding what sits on the shelf
Neither approach is without merit. Here is how forecasting-assisted planning differs from methods based mainly on experience and habit.
Back to homeWhy a comparison is useful
Most retailers already have a working method for reordering, built up over years of watching what sells. Forecasting tools do not replace that judgment — they give it a second reference point. Understanding where the two agree and where they diverge is often more useful than treating one as simply correct.
Traditional approach vs. forecasting-assisted approach
| Aspect | Traditional (experience-based) | Forecasting-assisted |
|---|---|---|
| Reorder trigger | Fixed threshold set once, rarely revisited | Threshold checked against recent movement and lead time |
| Seasonal effects | Adjusted by memory of past years | Modelled using calendar patterns specific to Japan |
| Confidence | Rarely stated explicitly | Shown as a margin of error alongside each figure |
| Time required | Low ongoing effort once set | Some setup time, then a light weekly review |
What sets this approach apart
The models used are configured around your own transaction history rather than a generic template, and they are tuned to calendar effects that matter locally — long holiday periods, seasonal changeovers, and event-driven spikes that a generic system might miss.
Every forecast is paired with an estimate of how much confidence to place in it, so the output is a starting point for a decision rather than an instruction to follow without question.
On effectiveness
Forecasting models tend to reduce the gap between what is ordered and what actually sells, particularly for lines with steady, well-established demand. The improvement is generally more visible in reduced stockouts and reduced excess stock than in any single dramatic change.
Results vary by category, and the caution bay on our home page outlines where this approach is less reliable — new lines and irregular, event-driven items in particular.
Cost and benefit, considered together
Each engagement has a fixed price and scope, set out clearly before work begins. The return is typically measured in reduced holding cost and fewer missed sales rather than in any single number, and it accumulates over the weeks following delivery rather than immediately.
What working together looks like
Compared with an internal, informal process, working with us adds a defined scope, a fixed timeline, and a written explanation of the assumptions behind each figure. You retain the existing decision — the forecast informs it rather than replacing it.
Sustainability of results
A reorder table or forecast configuration is not a one-time fix. Its usefulness depends on periodic review as your range and sales patterns shift, which is why each engagement includes a written note on assumptions that can be revisited later, with or without our continued involvement.
A few common misconceptions
Forecasting replaces the buyer's judgment
It does not. The output is a figure to weigh alongside what you already know about the store and the range.
Traditional methods are outdated
Experience-based methods work well for many retailers, particularly with stable, well-known ranges. The two approaches are not mutually exclusive.
More data always means a better forecast
Beyond a certain point, more history mainly helps with confidence, not accuracy — recent patterns often matter more than volume of history.
If you'd like to see how this applies to your range
A short note describing what you currently track is a reasonable place to start.
Get in touch