What Drives Your Demand
Four or five variables explain most of the variation in most operations. Finding yours takes an afternoon and improves every forecast after it.
Demand feels unpredictable until somebody looks. In most operations a small number of drivers account for the large majority of the variation.
The planning problem in “What Drives Your Demand” becomes clearer when scheduled hours can be compared with actual project and time records. Teams researching this workforce solution for attendance sheet template can add that operational evidence, while demand, service levels and manager judgement remain necessary to explain why a variance occurred.
The usual drivers
Day of week, which is the largest in nearly every business.
For an independent reference related to “What Drives Your Demand”, consult the CIPD workforce-planning resources; it provides a useful external check on scheduling, working-time and workforce-planning assumptions.
Time of day.
Season or month.
Weather, in some operations decisively.
Paydays, benefit dates, term dates, local events.
And anything you do yourself: promotions, opening hours, marketing.
Finding yours
Take a year of demand data, by day.
Plot it against each candidate driver in turn.
Most operations find that two or three explain most of the shape, and the rest is noise.
An afternoon, and the data is usually already in the till, the booking system or the case log.
The self-inflicted drivers
A promotion, a price change, an opening-hours change, a local campaign.
These are knowable in advance and are frequently not told to the person building the schedule.
Which produces a predictable miss that was entirely avoidable — and fixing the information flow is cheaper than any forecasting improvement.
The ones you cannot control but can know
Local events: a match, a festival, a conference, a school holiday.
A calendar of these, maintained once, removes a category of surprise.
Most operations rediscover the same events every year.
Weather, where it matters
In hospitality, retail with outdoor trade, and some services, weather is a large driver.
A simple rule beats nothing: hot and dry, add this much; heavy rain, take this much off.
Derived from your own history rather than from intuition, because the direction is sometimes the opposite of what people expect.
What to do with the drivers
Build the forecast as a base shape plus adjustments.
Base: this day of week at this time of year.
Adjustments: the event, the promotion, the weather.
Simple, transparent, and improvable — which a single number pulled from memory is not.
When nothing explains it
Some demand genuinely is volatile: emergency work, inbound faults, walk-in trade in some settings.
Then the answer is flexibility rather than prediction, and the coverage section covers it.
Knowing that your demand is genuinely unpredictable is itself a finding, and it changes what you build.
What to check
Which two drivers explain most of your variation?
Does the person building the schedule know about promotions in advance?
Is there an events calendar?
And have you ever checked the weather effect against your own data?