Test the unit before you sell the system: a unit-economics checklist built from 2026 evidence

Start from the failure mode

Franchise systems rarely collapse because the brand was weak. They collapse because a single outlet could not earn enough for the operator to survive after paying rent, labour, food cost, royalty and the loan on the fit-out. Everything else — territory maps, marketing funds, training academies — sits on top of that one arithmetic question.

That is why unit economics comes before disclosure, before recruitment, before any development schedule. And the market is currently supplying unusually clear evidence about what happens when the arithmetic is wrong.

Four evidence anchors from 2026

The first anchor is what pruning reveals. VnExpress reported on 21 April 2026 that Mixue overseas network fell by about 428 stores during 2025, ending at 4,467 international locations, with reductions concentrated in Vietnam and Indonesia. In the same period revenue reached roughly RMB33.56 billion and net profit RMB5.93 billion, both up more than 33%. Most instructive: stores moved to a newer, larger format averaged roughly 1.7 times higher revenue.

Read that as a unit-economics statement. If relocating a store nearly doubles its revenue, the original problem was never the product or the brand. It was site selection and store design — two variables set before opening day and nearly impossible to fix afterwards.

The second anchor is capacity. VnExpress reported on 20 September 2026 that Golden Gate opened a food plant in Phu Tho on 19 September 2026, costing more than VND600 billion, with the chief executive saying it lifts operating capacity fourfold, to the equivalent of about 2,000 restaurants, against roughly 600 today. A central plant moves prep cost and quality variance out of the store. That changes the unit model: lower in-store labour, more consistent food cost, higher fixed cost at the centre paid for by volume.

The third anchor is market growth rate. According to the iPOS.vn report produced with Nestlé Vietnam and validated by VIRAC, reported by Bao Dau Tu on 8 April 2026, Vietnam is forecast to have about 333,600 F&B outlets in 2026 on revenue of roughly VND760,000 billion, with revenue 4.6% above the VND726,500 billion recorded in 2025 while outlet numbers rise only about 1.2% from 329,500. Meanwhile VTV reported on 21 September 2026 that fast-food outlets grew 13% year on year in the first half of 2026.

A market adding about 1.2% outlets while one organised segment adds 13% is a market where new units are mostly taking share, not riding growth. That has a direct consequence: sales forecasts for a new outlet should be built from the competitive set within the catchment, not from national market growth.

The fourth anchor is pace discipline. Tuoi Tre reported on 8 May 2026 that bhc entered Vietnam through a master franchise with Hao Open Foods targeting 50 stores in ten years — five a year. Slow, published targets usually mean the operator intends to prove the unit before scaling it.

The checklist

Work through these before selling a single franchise. The calculator on this site takes the same inputs.

• Prove the unit twice, in different site types. One strong flagship proves the founder. Two profitable outlets in different catchments — one prime, one secondary — prove the model.
• Model the outlet without the founder. Add a full-cost store manager salary and remove any unpaid family labour. If the outlet stops being profitable, you have a job, not a franchise.
• Separate the four cost blocks. Rent as a percentage of sales, labour as a percentage of sales, cost of goods, and the amortised fit-out. Each needs its own ceiling, because each fails differently.
• Set the payback period before setting the fee. If a franchisee cannot recover the fit-out and franchise fee within a defensible period from unit cash flow, the royalty rate is irrelevant — nobody good will renew.
• Stress the rent line. Retail rent is the variable most likely to move against you over a five-year term and the hardest to renegotiate.
• Budget for relocation and closure from day one. Mixue 1.7 times finding argues that moving a store is a normal cost of running a network, not an admission of failure. Say in the agreement who pays for it and at what performance threshold.
• Separate central capacity cost from unit cost. If a plant or commissary carries fixed cost, state the volume at which it breaks even, so growth targets are grounded rather than aspirational.

A note on royalty design

A royalty is a claim on revenue, while the franchisee survives on cash after fixed costs. In a market where units compete for share rather than ride growth, a royalty set from optimistic revenue assumptions quietly transfers risk to the operator. Modelling the franchisee cash position at 70% and 85% of forecast revenue, not only at plan, is the cheapest diligence available.

What we do not know

• None of the 2026 announcements above disclosed franchise fees, royalty rates or per-store investment, so no external benchmark for these figures can be drawn from them.
• The Mixue 1.7 times figure comes from company reporting via VnExpress. The underlying store-level data and the definition of the comparison set are not public.
• Golden Gate capacity equivalent to 2,000 restaurants is designed capacity, not a commitment to open that many, and the cost allocation per unit is not disclosed.
• We could not find official Vietnam data published this week on closure rates among franchised outlets, which is the single most useful benchmark for setting a realistic survival assumption.

This article is educational analysis of publicly available information. It is not investment, legal or accounting advice, and it promises no level of return.