AI services differ on franchise suggestions: ChatGPT names no brands in 83% of responses

Ask generative AI to suggest franchises, and the nature of the answer will depend on the service you use. In its October tracking study, published on 5 November 2026, Branding Works found that ChatGPT named no specific brands in 83% of responses. Gemini, meanwhile, mentioned Osouji Honpo in 73% of responses. For anyone considering entering Japan’s franchise market, the findings underline the importance of looking beyond brand names to the assumptions behind an answer.

The same question can produce suggestions or requests for more information

The study repeatedly asks ChatGPT and Gemini, “Which franchises would you recommend?”, and records the company names appearing in their responses. Branding Works uses an in-house measurement tool to log generative AI responses daily. This release covers the results for October 2026.

By service, Lawson had the highest mention rate on ChatGPT, at 13%, while Osouji Honpo led on Gemini, at 73%. These figures cannot, however, be directly compared as measures of how highly each brand is rated. Most ChatGPT responses did not reach the stage of naming brands at all.

According to the release, ChatGPT frequently asked follow-up questions about the user’s circumstances or explained general principles for choosing a franchise. Not naming a brand is different from giving it a poor assessment. The notable distinction in this study is therefore not just which candidates each service suggests, but whether it first seeks to establish the user’s requirements.

Combined rankings can conceal differences between services

Across the study as a whole, Osouji Honpo ranked first, with a mention rate of 36.5% and an average position of second in responses. Kaitori Daikichi followed at 31.1%, then Otakaraya at 30.5%. However, all three brands were named only by Gemini and did not appear in ChatGPT’s responses.

A high overall ranking therefore does not mean that both AI services suggested the same brand. When reading figures that combine multiple services, it is important to check which AI’s responses account for the result.

The company defines the mention rate as the proportion of all responses in which a company’s name appears. Average position refers to the order in which names appear within a response. Where mention rates are tied, average position determines the ranking.

The mention rate thus measures how often a name appeared in response to this particular question. It does not measure franchisee profitability, the quality of support from the franchisor or suitability for a prospective franchisee. The starting point for using these figures is to avoid equating a frequently mentioned brand with one that meets your own requirements.

Magokoro Bento rises by 15.1 percentage points from the previous month

The top three brands were unchanged from the previous month. Magokoro Bento recorded the largest increase in mention rate, rising by 15.1 percentage points from 13% in September to 28.1% in October. It ranked fourth overall in October, with an average position of 6.5 in responses.

Duskin and IBJ entered the top ten, while Life Deli and 7-Eleven dropped out. Duskin and IBJ both had mention rates of 12%, with average positions of 3 and 4.9 respectively.

These results show that the mix of brands appearing in generative AI responses can change from month to month. However, a rise in mention rate cannot be interpreted as an increase in franchise applications or outlet numbers. The findings show changes in AI responses, not in franchise recruitment performance.

Magokoro Bento’s figures also clearly illustrate that mention frequency and position within a response are different measures. A brand may appear in many answers without consistently being listed first. When reading a ranking table, it is important to consider the definitions of the measures rather than relying on a single figure.

Treat AI answers as a starting point for your franchise search

The results highlight why prospective franchisees should not settle on a shortlist based solely on one AI service’s answers. Brands frequently mentioned by Gemini did not appear on ChatGPT, while ChatGPT often sought to clarify the user’s requirements. Simply checking whether a name appears does not fully capture that difference.

Before asking AI for suggestions, prospective franchisees could set out their available investment, preferred location and the time they can devote to running the business themselves. They should then check the franchisor’s published entry requirements, fees and support for any brands suggested, and confirm the details through information sessions or meetings.

Franchisors reading this type of study also need to look beyond the overall ranking. They should distinguish between mention patterns on each AI service and responses that do not name brands at all. Where a service frequently asks about the user’s requirements, as ChatGPT did in this study, a low mention rate cannot automatically be attributed to insufficient communications or publicity.

The practical takeaway is not to treat an AI-generated shortlist as a final decision on which franchise to join. Check what the figures measure and the assumptions behind the answers. Once you have identified candidates, compare information supplied directly by the franchisor with your own operating requirements.