Most financial models built for an Indian consumer business have a single line item somewhere in Q3 or Q4 labelled “festive season,” treated the way a US company might treat Black Friday or the Christmas week: a predictable bump, roughly the same size, roughly the same time, every year. That model is wrong in a way that’s easy to miss until it costs a company real money, because India doesn’t have one festive season. It has a rolling sequence of distinct regional demand waves stretched across nearly three months, most of them following a lunar calendar that refuses to sit still on the same Gregorian dates two years running.
It is a season, not a day, and it moves
Start with the scale, because it is large enough to change how a company should think about its entire year. The 2025 festive trading period generated an estimated Rs 6.05 lakh crore in retail trade, up 25% from the previous year, and the run-up to Diwali alone drove roughly Rs 1.15 lakh crore in e-commerce GMV over just 30 to 35 days, with order volumes up 24% year on year and quick commerce order volumes more than doubling. That period wasn’t a single week. It ran across roughly 87 days, from early August to late October in 2025, and inside that window sat Onam, Ganesh Chaturthi, Navratri, Durga Puja, Dhanteras, and Diwali, each one landing hardest in a different part of the country.
Onam belongs to Kerala. Ganesh Chaturthi is concentrated in Maharashtra and Karnataka. Durga Puja is a public holiday specifically in West Bengal, Odisha, Tripura, Bihar, and Assam, and Kolkata’s Durga Puja and Kali Puja trade alone was estimated at close to Rs 17,000 crore in a single recent year. Pongal belongs to Tamil Nadu and arrives months apart from all of this, in mid-January. A company that plans for one national festive spike is really missing five or six separate regional ones, each requiring its own inventory position, its own marketing calendar, and often its own product mix, since what sells during Onam in Kerala looks nothing like what sells during Ganesh Chaturthi in Pune.
The dates themselves are not fixed, and that is the part planning systems miss
The harder problem sits underneath all of this. Most of these festivals follow the Hindu lunisolar calendar, which is pegged to the moon rather than to a fixed point on the Gregorian calendar most business planning tools are built around. Because a lunar year runs roughly eleven days shorter than a solar one, festivals drift earlier each year until a periodic calendar correction pulls them back, which is why Diwali can land anywhere from mid-October to mid-November depending on the year. That is not a rounding error. It is enough movement that a festival which fell inside a company’s Q3 planning window one year can fall inside Q4 the next, and a business that simply copies last year’s promotional calendar forward by twelve months, rather than re-deriving the actual dates, will find itself either weeks early or weeks late to its own biggest revenue period.
This is exactly the kind of detail that looks minor from a strategy deck and turns out to be operationally significant, because inventory has to be positioned well ahead of the peak, not during it. A meaningful share of manufacturing, logistics, and sourcing partners effectively slow down or close entirely for the week surrounding the festival itself, right at the moment demand is highest, which means the actual planning window a company has to get stock into place is shorter than the headline season length suggests.
Weddings stack another calendar on top of this one
For a meaningful set of categories, jewelry, apparel, gifting, hospitality, and event services, there is a second calendar layered directly on top of the festival one, and it behaves even less like a smooth season. India’s wedding season runs broadly from October through March across much of the north and west, with a second major spike around Akshaya Tritiya in April or May, widely considered the single most sought after wedding date of the year. Within that broader window, demand doesn’t spread evenly either. It concentrates on specific auspicious dates, muhurats, determined by the Hindu calendar, so thin a set of dates that thousands of weddings can land on the same day in the same city, which is why venues, jewelers, and hospitality vendors see demand arrive less like a wave and more like a series of sharp spikes on specific dates that shift every year along with the rest of the lunar calendar.
What this actually means for how a company should plan
None of this is an argument for treating the festive period as unpredictable. It is quite predictable, years in advance, in the sense that the underlying calendar is fully calculable. The mistake is treating it as one event that can be planned for with one number and one set of dates copied from the year before. A business operating pan-India is effectively running five or six separate regional campaigns in sequence, each tied to a date that needs to be recalculated every year rather than assumed, each requiring inventory to be in place before local logistics capacity tightens, and each demanding a product and marketing mix suited to that specific region rather than a single national push dressed up as local.
The businesses that handle this well tend to build their planning calendar around the actual lunar dates for the specific year in question, region by region, well before the quarter starts, rather than defaulting to whatever slide worked last time. Given how much of the year’s revenue can concentrate into this stretch for a consumer business in India, getting that sequencing right is not a seasonal nicety. It is close to the central planning problem of the operating year.
