A new site can look strong on traffic and still be a bad idea, because some of the sales it projects aren't new sales at all. They're guests who already eat at the restaurant down the road, driving a bit further than they used to. The only way to see this is to map where your current guests live and put a dollar figure on what moves.
How do you find out how many guests a new site would take from an existing location?
Map where your delivery orders and guest list live relative to the new site, then estimate what share of the first location's guests are closer to the new one and would switch.
Delivery orders and a guest list already show where people who eat at the first location actually live. Plotting that against a candidate site shows how much overlap exists. A site a mile from an existing one will share more guests than a site across town. The map comes before the lease, not after.
What does it look like when a new site takes guests from the first restaurant, in dollars?
In the lesson's worked example, thirty percent of the first location's guests live closer to a candidate site, half of them switch, and that fifteen percent of sales works out to ninety thousand dollars a year of profit moved rather than created.
The arithmetic in the example: if thirty percent of the first location's guests live closer to Site C and half of them switch, that's fifteen percent of the first location's sales. On a million five hundred thousand dollar base, that's two hundred and twenty-five thousand dollars of sales. At forty cents kept per dollar, that's ninety thousand dollars a year. Site C, in that example, doesn't add a location. It moves one.
Does even a far-away site take some guests?
Yes, in the worked example even the site with the least overlap still took a small, countable share.
In the example, Site A takes only two and a half percent of the first location's sales. At forty cents a dollar, that's fifteen thousand dollars a year. So instead of counting the full hundred and fifty thousand dollars the new site was projected to make, the group counts what it actually gains: a hundred and thirty-five thousand dollars. Against a six hundred thousand dollar project, that changes payback from forty-eight months to fifty-three months. Small, but it belongs in the pro forma, not left out.
Results are not typical and will vary with your business, your market, your team and how much of the work you actually do. Client figures on this site come from recorded interviews and are dated. Nothing here is a guarantee of revenue, profit or growth.
Why does this matter more than how good the site looks on paper?
Because a site can score well on traffic and visibility and still be a net loss for the group once the guests it takes from an existing location are subtracted.
A location that scores well on five lines and poorly on overlap with your own guests isn't a new stream of sales. It's a transfer, with build costs attached. The six-line scoring system used before a letter of intent treats this as one of the deciding lines, alongside rent against projected sales, because the two together sink or save most sites before they ever reach a pro forma.
What should a restaurant owner do before signing a lease on a new site?
Run the overlap numbers alongside rent against projected sales, score the site on both, and only move to a letter of intent once the group's actual gain, not the site's standalone projection, clears the bar.
Someone stands on the site at the busy hours that matter: lunch on a weekday, dinner on a Friday. They watch who eats nearby and compare that to a demographic report. That's the traffic and the local-guest lines. The guest-overlap line comes from the existing restaurant's own data. Put the two together in dollars, and a site that looked like the best one on the sheet can turn out to be the one that sinks the slowest.
The scoring and the arithmetic behind it are part of the Scaling Engine OS, the system licensed to members of the Founders Board. Owners who've put a second or third location through it are interviewed on case studies, including one who took four months to sign a second lease after opening the first.
Can rent alone sink a site even if guest overlap looks fine?
Yes, rent against projected sales is the other line that decides most sites, and it has to be checked in dollars, not just as a percentage against traffic.
A site with the best traffic on a shortlist can still carry a payback near ten years once its rent is run through the full cost cascade, while a site with worse traffic and lower rent pays back in under half that time. Guest overlap and rent are the two lines worth turning into dollars before anything else on the sheet.

