What the Fastest Retail Operators Do Differently

The gap between operators is wider than capital

Line up the US retailers and restaurant brands that open the most stores and a pattern shows up fast. Dollar General opened 581 new stores in the United States in fiscal 2025, on real estate plans that run into the hundreds of new builds a year. Chipotle put up 304 in 2024, with 257 of them carrying a Chipotlane. Chick-fil-A added a net 179 outlets in 2025 to reach 2,863. Dutch Bros opened 154 shops, and Raising Cane's set a company record at 118.

A rural dollar store, a fast-casual burrito line, a chicken drive-thru, a coffee chain, and a chicken-finger concept have almost nothing in common at the register. Under the hood they build the same way. Plenty of better-funded chains open a fraction of these totals, so the money clearly is not doing the work. What these operators share is a small set of decisions they make before a single site goes under contract, and every one of them holds up just as well at 10 stores a year as at 500.

One store, designed once and built hundreds of times

It starts with a controlled prototype. Dollar General runs most of its new stores on a single 8,500-square-foot format, which lets it plan cost, crew, and schedule against a known baseline instead of starting fresh every time. Chick-fil-A goes further and builds a standardized freestanding drive-thru that the company itself designs, sites, and equips before handing the keys to an operator. Chipotle made the Chipotlane its default growth format and now opens the large majority of new restaurants with one. Dutch Bros has leaned into build-to-suit, where a developer delivers a shell built to the brand's spec.

The point of a locked format is that the team stops arguing about the building on every project. Your estimate turns into a lookup instead of a negotiation. Permitting barely changes between jurisdictions once the drawings are the same store, and any contractor who built the last one already knows what the next one takes. Variance is what slows a program down, and a standardized box has removed most of it before design even begins.

What a real prototype standardizes

The discipline is in holding the line on one format, because it earns the name prototype only when it fixes the decisions that would otherwise get re-argued at every location.

• Format and square footage, so every site works from one baseline instead of a custom design.
• An FF&E and equipment package a vendor can quote once and stock ahead of the pipeline.
• A repeatable permit and drawing set that changes little from one jurisdiction to the next.
• An assigned or program general contractor who has built the format before and prices it fast.
• A cost-per-square-foot target the estimate is measured against, not discovered after the fact.

Owning the development function instead of renting it

Speed like that needs a real development organization, staffed and accountable, that owns the work rather than passing it between operations and finance. Chick-fil-A is the clearest case. The company controls site selection, construction, and equipment for its restaurants and keeps ownership of the assets, so the discipline lives in one place instead of scattering across hundreds of franchisees.

Raising Cane's runs its pipeline like a production line. The company told Fox Business it keeps roughly 300 restaurants at various stages of development at any given time, feeding a long-term goal of more than 1,600 locations. That standing inventory of sites exists because someone owns the growth number and builds a pipeline deep enough to hit it, instead of reacting to whatever a broker sends over. Chipotle has described building a real estate pipeline deep enough to sustain 8% to 10% unit growth for years. Velocity like this is manufactured, the output of a team whose whole job is keeping sites moving.

Cost and schedule as the product

Then there is measurement with teeth. Dollar General's CEO told investors that a new store now averages about $500,000 to open, including capital and inventory, that build cost has climbed more than 40% since 2019, and that the company gates against a portfolio return target of roughly 17%. That is an operator who knows its per-store cost to the dollar and will hold a site back when it misses the return threshold. Dutch Bros has shown the same instinct, crediting its shift to build-to-suit for pulling capital expenditure per shop down quarter over quarter.

Holding a schedule is getting harder as the labor market tightens. In the Associated General Contractors of America's 2025 workforce survey, 92% of firms reported difficulty filling positions and 45% said worker shortages were already delaying projects. When crews are scarce, hitting an opening date comes down to tracking permitting and long-lead equipment as named risks on every job. A fixed prototype and a program contractor are what make that tracking possible, because everyone involved already knows what the building takes.

What this means for a 10-store-a-year program

None of this needs Dollar General's volume behind it. A 10-store-a-year operator can pick a prototype and hold the line on it, so estimates and permit sets stop starting from zero. Put one person on the pipeline. Give them a real growth number and a standing list of sites to feed it, and hold every project's actual cost per square foot and weeks-to-open against the prototype baseline. When a format drifts more than 10% to 15% above where it should sit, re-baseline the budget before the next store starts.

The hard part at any scale is keeping that data somewhere a leadership team can see it. When per-store cost lives in one analyst's spreadsheet and the schedule lives in a thread of forwarded emails, the discipline that drives velocity erodes without anyone deciding to let it. A purpose-built store development platform is one way to hold the baseline steady across a portfolio, though the discipline matters more than the tool. For the fastest operators, how they build is a settled question well before they start looking at where the next store goes.