Size and presence
In September franchise networks open 20% more units per business day than the annual average, and in January, 23% fewer
Openings of network units per business day in September, relative to the annual average, from 2015 to 2025, in 61,722 establishments
See the full table (9 rows)
| Segment | CNPJs opened | Highest month | Lowest month | Index in the highest month | Other activities in the same month | Correlation between years |
|---|---|---|---|---|---|---|
| Food | 14,204 | October | January | 117 | 106 | 0.56 |
| Health, beauty and wellness | 12,550 | September | December | 116 | 107 | 0.43 |
| Services and other businesses | 11,891 | August | January | 116 | 107 | 0.49 |
| Fashion and accessories | 10,730 | September | January | 136 | 113 | 0.72 |
| Food, retail and distribution | 6,660 | October | January | 141 | 106 | 0.65 |
| Education | 3,087 | no stable pattern | no stable pattern | n/a | n/a | 0.18 |
| Gyms and sports | 2,194 | no stable pattern | no stable pattern | n/a | n/a | -0.13 |
| Tourism and hospitality | 406 | no stable pattern | no stable pattern | n/a | n/a | 0.01 |
| All measured networks | 61,722 | September | January | 120 | 109 | 0.74 |
Monthly index by segmentOpenings per business day, with the annual average equal to 100, from 2015 to 2025. Segments without a stable pattern across years (education; gyms and sports; tourism and hospitality) are left out.
| Subset | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Food | 75 | 103 | 96 | 94 | 94 | 96 | 105 | 112 | 114 | 117 | 112 | 83 |
| Health, beauty and wellness | 79 | 97 | 103 | 97 | 91 | 107 | 99 | 114 | 116 | 114 | 107 | 75 |
| Services and other businesses | 79 | 92 | 100 | 106 | 94 | 97 | 104 | 116 | 114 | 107 | 108 | 83 |
| Fashion and accessories | 64 | 85 | 93 | 90 | 90 | 95 | 111 | 116 | 136 | 130 | 119 | 69 |
| Food, retail and distribution | 70 | 94 | 96 | 79 | 88 | 92 | 107 | 114 | 127 | 141 | 120 | 71 |
| All measured networks | 77 | 97 | 98 | 94 | 93 | 98 | 104 | 113 | 120 | 118 | 112 | 78 |
| Other establishments in the same activities, standardized | 83 | 109 | 105 | 97 | 98 | 101 | 104 | 108 | 109 | 105 | 103 | 79 |
The same index in check subsetsProfile of all measured networks recalculated in each subset, with the annual average equal to 100.
| Subset | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without 2020 and 2021 | 75 | 93 | 99 | 100 | 98 | 100 | 105 | 112 | 118 | 115 | 107 | 77 |
| Without the growth trend | 79 | 99 | 100 | 95 | 94 | 98 | 104 | 112 | 118 | 115 | 108 | 75 |
| Without batch openings by the same company | 77 | 96 | 98 | 94 | 94 | 98 | 105 | 112 | 119 | 118 | 113 | 78 |
| Without each network's first openings | 77 | 96 | 97 | 93 | 93 | 98 | 104 | 113 | 120 | 119 | 112 | 78 |
| Head offices only | 78 | 98 | 99 | 94 | 94 | 100 | 103 | 111 | 118 | 115 | 112 | 77 |
| Branches only | 72 | 90 | 94 | 89 | 90 | 88 | 113 | 121 | 123 | 126 | 110 | 80 |
| Without the largest network | 77 | 96 | 98 | 95 | 94 | 98 | 104 | 113 | 119 | 116 | 111 | 79 |
| Only establishments active today | 75 | 95 | 94 | 88 | 91 | 99 | 103 | 115 | 123 | 123 | 114 | 79 |
| Only establishments not active today | 84 | 102 | 104 | 104 | 101 | 95 | 103 | 109 | 114 | 106 | 105 | 74 |
| Networks with equal weight | 81 | 100 | 95 | 98 | 94 | 99 | 103 | 111 | 118 | 117 | 108 | 76 |
The start-of-activity date in the CNPJ (Brazilian company registry) records makes it possible to observe in which month of the year franchise networks register new units. The indicator counts 61,722 establishments, each corresponding to a 14-digit CNPJ with the brand in the trade name, opened from 2015 to 2025 in the 502 measured networks with openings in the period, and calculates an index per business day in which the average of each year equals 100.
The index reaches 120 in September and 77 in January. The period from August to November accounts for 38.2% of openings, compared with 33.6% of business days and 34.6% of the openings of other establishments in the same activities. September was above average in 11 of the 11 years, and January in none. The profile of each year has an average correlation of 0.74 with that of the other years, with the lowest values in 2020 and 2025.
Among other establishments in the same CNAE (national classification of economic activities) codes, weighted by the networks' mix of activities, the distance between the highest and the lowest month is 30 points, compared with 43 in the networks, with an index of 109 in February and 109 in September. The networks' index exceeds that of the other activities from September to November, repeatedly in at least 9 of the 11 years. By segment, there is a stable pattern in food; in health, beauty and wellness; in services and other businesses; in fashion and accessories; and in food, retail and distribution, with the month of highest registration between August and October. In education, in gyms and sports and in tourism and hospitality, there is no stable pattern from one year to the next, and the page does not attribute a peak month to these segments.
How we calculate
We started from the establishments that the observatory's census attributes to the measured networks, by the trade name expression and the CNAE (national classification of economic activities) subclasses of the network profile, in any registration status in the reference month of October 2026, and took the month of the start-of-activity date. For each year from 2015 to 2025, the index of a month is the ratio between openings per business day in that month and openings per business day in the whole year, multiplied by 100, excluding weekends and national holidays, including Carnival and Good Friday. The published profile is the average of the annual indices. Records dated January 1st, a possible loading artifact, total 2 in the networks and 387 in the other activities and do not change the result.
The comparison uses the other establishments with a trade name in the same CNAE codes, outside any CNPJ root of the networks, with the index calculated by CNAE code and weighted by the share of each code in the segment's networks. Companies that have opted into the MEI (individual microentrepreneur regime) are excluded from both sides, 4,197 establishments in the networks, because MEI registration has its own calendar, with an index of 122 in January, and because in the networks this case tends to indicate a CNPJ that predates entry into the brand.
Stability is tested year by year by the correlation of each year's profile with the average of the other years, and by permutation of the months within each year. A segment only has a peak month published if the average correlation is at least 0.4, if the permutation test rejects chance at the 1% level, if at least 9 of the 11 years have a positive correlation and if the result holds without the segment's largest network.
The high season comprises the months with an index above 100 in at least 8 of the 11 years. So that a single year's fluctuation at the threshold does not change the period from one edition to the next, a month in the period published in the previous edition only leaves it when it is above the average in 6 years or fewer, and a new month only joins it with 8 years or more. In this edition, November remains in the period with 7 years above the average.
What this figure does not show
- The date is that of the CNPJ registration, not of the inauguration. The interval between the two events varies and is not observed in the base.
- A unit that inherited an old CNPJ, through the conversion of an existing business, keeps the original date. This case cannot be reliably identified in the base and tends to bring the networks' profile closer to that of the other activities. Without the openings before the tenth percentile of each network's dates, the September index is 120.
- The trade name is the current one. Units that left the network and changed their name drop out of the count, namesakes are included and units with another name are left out. Among establishments active today the September index is 123, and among those not active today, 114.
- The profiles of 2020 and 2021 are affected by the pandemic. Without these years, the September index is 118, and after removing the annual growth in openings, 118.
- In food, retail and distribution, the largest network accounts for 53.3% of openings, and without it the correlation between years goes from 0.65 to 0.48.
- The indicator describes a calendar of registrations and does not identify its cause. The comparison includes independent businesses and does not measure an effect of the franchise model.
Frequently asked questions
In which month do franchises open the most units?
Based on the start-of-activity date in the CNPJ (Brazilian company registry), from 2015 to 2025, the peak is September, with an index of 120 per business day, for an annual average equal to 100. The period from August to November accounts for 38.2% of the openings of the measured networks, compared with 33.6% of business days.
Does January have more franchise openings than other months?
No. In the measured networks, January has an index of 77 per business day, for an annual average equal to 100, and was not above average in any of the 11 years. MEI (individual microentrepreneur regime) registration in the same activities follows another calendar, with an index of 122 in January.
Do franchise openings follow the same calendar in all segments?
No. There is a stable pattern in food; in health, beauty and wellness; in services and other businesses; in fashion and accessories; and in food, retail and distribution, with the month of highest registration between August and October. In education, in gyms and sports and in tourism and hospitality, the month of highest registration changes from one year to the next, and there is no stable pattern.
Report an error on this page (in Portuguese)
Size and presenceOther indicators in the same group
Maintained by Marktech, which serves franchise networks and franchises. The brands cited belong to their owners and are mentioned here only to identify the network the data refers to.