A new-store sales forecast answers the one question every expansion hinges on: how much will this address actually sell in year one? Done well, retail sales forecasting replaces the optimism baked into a broker’s pitch with a number you can underwrite. It de-risks site selection by pricing the downside before you sign, so a shaky location fails on a spreadsheet instead of on your P&L.
A new-store sales forecast is a modeled estimate of a prospective location’s first-year revenue, built from trade area, demographics, foot traffic, competition, and your own store performance. It converts a gut-level “this feels right” into a defensible range you can test against rent and returns.
WHAT IT IS
A forecast is a de-risking tool, not a crystal ball
A new store sales forecast estimates first-year revenue for a location you have not yet leased. Its job is not to be exactly right — it is to be reliably directional, so you can rank sites, size the opportunity, and walk away from the ones that only work in a best case. The forecast becomes the spine of the pro forma: it sets the sales line that rent, labor, and payback all get tested against. A location that needs everything to break its way to clear its rent is a location the forecast should surface before you sign.
THE INPUTS
What actually feeds a credible number
Good revenue forecasting blends several layers. The trade area defines the catchment; demographics and psychographics describe who lives and works inside it; foot traffic and mobility data show real movement past the door; competition maps who else is fighting for the same wallet. Layered on top: comparable or analog stores and, most valuable of all, your brand’s own historical performance — the pattern of what drives sales across the units you already operate.
THE METHODS
Analog, regression, and machine learning
The simplest approach is analog: find existing stores that look like the candidate and borrow their sales. It is intuitive but brittle — no two trade areas match cleanly, and analogs quietly smuggle in the biases of whoever picked them. Regression improves on this by weighting the variables that statistically move sales. Machine-learning models go further, capturing non-linear interactions — how income, competition, and traffic combine rather than add — and learning from every new store you open. Each method is a step away from anecdote and toward evidence.
WHY AI WINS
Why predictive analytics beats gut feel
Predictive analytics in retail wins because it is consistent, testable, and honest about uncertainty. Gut feel and a single hand-picked analog reward confirmation bias; a model applies the same logic to every site and shows its work. It weighs dozens of variables no operator can hold in their head at once, and it improves as your portfolio grows. Crucially, it separates a genuinely strong site from one that merely resembles a favorite store. See where this is heading in the future of retail site selection.
ACCURACY & USE
Confidence ranges and the go/no-go
A forecast should arrive as a range, not a single number, with the confidence band widening for unusual sites and narrowing where you have dense comparable data. Validation matters: hold-out testing against actual results and back-testing on stores you have already opened tell you whether to trust it. Then it does real work — it drives the go/no-go, sets the rent you can afford as a percentage of sales, and hands you a defensible figure at the negotiating table. Pair it with cannibalization analysis so a new unit’s forecast nets out sales it pulls from existing stores.
A forecast that only works if everything breaks your way is not a forecast — it is a wish with a spreadsheet.
Population counts alone will fool you — two trade areas with identical headcounts can spend very differently. Layer psychographics, daytime population, and spending behavior on top of the raw numbers. Start with demographic insights for site selection before you trust any forecast built on them.
Bottom Line
The bottom line
A new-store sales forecast is the cheapest insurance in expansion. It costs a fraction of a single month’s rent and it stops you from signing a decade-long lease on a location that will never clear its cost of occupancy. The goal is not perfect precision — it is a defensible, ranged estimate that ranks your options, sizes the risk, and gives you something concrete to negotiate against. In a market where every good site has competition, the brand that forecasts is the brand that signs the right leases and walks from the wrong ones.
Locate builds first-year revenue forecasts for every candidate site, combining trade-area analytics, foot traffic, and your own store performance in predictive models — then pressure-tests each number against rent so the go/no-go is grounded in evidence, not optimism.
FAQ