Piloting a franchise: how to test the model before selling to partners

A successful company-owned outlet does not, on its own, prove that the format offered to a franchisee will work. A franchise pilot tests a specific proposition under conditions close to those a future franchisee will face, using the planned equipment, cost structure and level of support. A sustainable franchise network needs repeatable results, not a showcase launch. Here is how to organise that test before scaling up.

1. Define the hypothesis the pilot will test

Do not simply aim to ‘make sure the franchise is good’. Set out a testable assumption: for example, that a compact outlet with a smaller product range can meet service standards and cover its costs on normal purchasing terms. The pilot should test the exact format you intend to offer franchisees, not a more expensive version or one receiving preferential treatment.

Prepare a pilot brief covering:

• the format: floor area, product range, equipment and opening hours;
• site conditions and customer traffic characteristics;
• the instructions and tools available to the manager;
• the planned level of support from the central team;
• performance measures and criteria for stopping the test.

Choose a site that matches the future offering. A flagship outlet with a long-established customer base is a poor testing ground for a new, smaller format. If you use an existing outlet, record separately any advantages that a franchisee will not have, such as an established customer base, discounted rent or shared warehousing facilities.

Set the observation period in advance. It should cover not just the opening, but also repeat purchases, stock replenishment, payroll and normal fluctuations in demand. There is no universally required pilot duration. For a seasonal business, peak-season results cannot be extrapolated to the whole year without qualification.

2. Separate the test conditions from the legal arrangements

You can run a pilot at a company-owned outlet or with an independent business owner. The first option is useful for early testing: it is easier to change processes and limit financial risk. However, it does not fully replicate the relationship between two independent parties. The second lets you assess whether the requirements placed on a franchisee are realistic and whether support is sufficient, but it requires the relationship to be properly formalised.

In Russia, commercial concession agreements—the legal framework used for franchising—are specifically governed by Chapter 54 of the Russian Civil Code. The intellectual property provisions in Part Four of the Code also apply. Calling an arrangement a ‘pilot’ does not exempt it from these rules. If an independent partner is granted, for a fee, a package of exclusive rights that includes the right to use a trade mark or service mark, you need to assess whether the commercial concession rules apply.

Under Article 1028 of the Russian Civil Code, such an agreement must be in writing. The grant of the right to use the package of exclusive rights must be registered with Rospatent, Russia’s intellectual property office. Without registration, the grant of rights is deemed not to have taken place. This should not be reduced to the claim that ‘every unregistered franchise is automatically invalid’. Commercial organisations and registered individual entrepreneurs may be parties to a commercial concession agreement.

If the pilot takes place within a single legal entity, there is no need to enter into a concession agreement with yourself. If the site is operated by a separate limited liability company (OOO), even one belonging to the same owner, the legal basis for its use of the brand must be checked separately.

Agree in advance with any independent participant how the costs of experimental changes will be shared. For example, who pays to replace equipment if the original specification proves unsuitable? Do not leave this risk covered only by a verbal promise to ‘sort it out later’.

3. Measure performance without hidden subsidies

The main mistake in a pilot is to count only the money passing through the outlet’s till. The central team may provide work free of charge that a future franchisee would have to pay for: running advertising, covering for a manager or delivering small batches of stock. In that case, an attractive result reflects the owner’s support rather than the strengths of the format.

Maintain two parallel calculations: the pilot’s actual performance and its performance under the proposed franchise terms. In the second, include the planned fees and the normal costs of purchasing, delivery, software and essential management functions. Label notional charges separately: they are for analysis, not for altering the accounting records.

Collect data in four areas:

• Financial performance: revenue, cost of sales, write-offs, payroll and cash balance.
• Operations: order fulfilment times, downtime, stock shortages and returns.
• Support: reasons for support requests, specialist time spent and recurring questions.
• Sustainability of demand: repeat purchases and performance after introductory promotions end.

Record every unplanned intervention: the problem, its cause, the action taken, the cost and the outcome. If a head office employee spends a week helping at the site, that is part of the cost of the test, not an invisible resource.

Do not change the product range, opening hours and advertising at the same time if you want to understand what caused an improvement. Where possible, test changes sequentially and keep a record of each version of the operating conditions. Otherwise, the final profit figure will not tell you which decisions actually worked.

4. Make the decision against criteria set in advance

Before the pilot begins, define the conditions for starting franchise sales. These should cover not just financial performance, but also operational feasibility, service quality and the demands placed on support. Set specific thresholds for your model rather than following a universal template.

There are three possible decisions at the end: the format is ready for a limited rollout, it needs further testing after changes, or it is not yet suitable for franchising. A negative result is useful if it prevents an unstable model from being passed on to franchisees.

Prepare a final report setting out what you tested, which conditions differed from those future franchisees will face, what costs you identified and what you changed. Distinguish confirmed observations from forecasts. One successful outlet does not prove that every location will deliver the same returns.

Practical takeaway: start selling franchises after a documented test of the format, not simply after an impressive pilot opening. You should understand the conditions in which it works, how much support it needs and which risks remain untested.