Skip to content
GullySales

Notes for owners · Business growth

How to promote a new product

Three jobs in order: awareness among people who already buy the category, a cheap way to try it once, and the shelf where they look actually having stock. Launch small enough to learn.

The GullySales team · Updated 15 Sept 2026 · 7 min read

A new product needs three things, in order. People who already buy the category have to hear about it, they need a cheap way to try it once, and the shelf they look at has to have it. Get the order wrong and you pay to create a want nobody can satisfy. So launch in one city, or one cluster of stores, rather than across a state. Judge the first quarter on the retailer's second order and the customer's second purchase. The rest is decoration until those two move.

Awareness, trial and the shelf

JobWhat does itWhat to count
Awareness among category buyersLocal creators, in-store branding, targeted social, the trade itselfReach inside the launch area, name searches
Trial, onceSampling, a smaller pack, an introductory price with an end date, demonstrationUnits tried, conversion from sample to purchase
The shelf, physicalDistributor and retailer coverage, placement at eye level, visibility at the counterStores billed, stores restocked
The shelf, digitalMarketplace and quick commerce listings, images, titles, first reviewsListing live, search rank inside the platform, reviews
RepeatPack, taste, function, and whether the store had stock the second timeSecond purchase rate, retailer reorder rate

Look at the fourth row before the first. A campaign that creates demand in an area where the product is not stocked sends your buyer to a competitor who is.

Launch small enough to learn

Pick an area you can drive around in a day.

For example, a millet snack brand launching in Bengaluru might choose 120 kirana stores and two supermarket chains across three or four localities. Add one quick commerce platform in the same pin codes. Small enough that the founder can visit twenty stores a week. Small enough that the whole launch stock fits in a tempo.

Inside that area, go deep. The retailer sees you every fortnight. The shelf strip is put up and replaced when it tears. The sampling happens on Saturday evening when families shop, not on a Tuesday afternoon. You hear the objection that a report would never carry, which is often about the price point or the pack size rather than the product.

Only after the repeat purchase holds for two months does it make sense to add the next city. Expansion before that multiplies a mistake across a state.

Trial is the expensive part, and worth it

Nobody buys an unknown brand at the same price as a familiar one without a reason. Trial is that reason, and it costs real money.

Sampling works where a person can judge quickly: food, personal care, anything with a taste, smell or feel. Put the sample in the hand at the point of purchase, with the product on the shelf an arm away. A sample given outside a metro station, three kilometres from any store that stocks you, is a donation.

A small pack works where sampling is impractical. It lowers the price of a first try and it earns a second shelf facing.

Demonstration works for anything mechanical or technical. A hardware brand launching a new tap fitting gets further by fitting one in the plumber's presence than by any advertisement, because the plumber decides what the household buys.

And the trade needs its own trial. A retailer who has never sold your category takes a risk with his shelf. A small first order, a clear margin, and a visit to check what moved is what turns him into a stockist rather than a one-time buyer.

The shelf, physical and digital

On the physical side, three things decide everything: whether you are in the store, where in the store, and whether stock is there when the customer comes back. Eye level and the counter are worth paying for. The back of the bottom rack is not worth having.

On the digital side, the listing is the shelf. The title has to carry the words people type. The first image has to be readable as a thumbnail on a phone. The pack size, ingredients and claims have to be exactly what the label says, because a mismatch is where returns and bad reviews come from.

Quick commerce has changed the timing of an Indian launch. A customer who sees a creator's reel at nine in the evening will look for the product on an app within a minute. If it is not listed in her pin code, the money you spent on that reel goes to whoever is.

The two numbers that tell you the truth

Count the retailer reorder rate and the customer repeat rate, weekly.

If retailers reorder and customers do not repeat, the product is being bought once and not again, which is a product problem no campaign fixes. If customers repeat and retailers do not reorder, the trade margin or the service frequency is wrong. When neither moves, stop spending on awareness and go and stand in a store for two hours.

Both numbers are available from your own invoices and, on marketplaces, from the platform's own reports.

What to do next

Write down the launch area, the number of outlets in it, and the date by which each will be stocked. Then plan the trial activity for the week after the stock lands, not before. If you would like the launch plan and the first quarter's measures worked out against your own distribution, book the free audit.

Questions

Questions owners ask.

Should a new product launch across the state or in one city?
One city, or one cluster of stores inside a city. A narrow launch lets you stand in the shop and watch people pick it up or put it back, which is information no report gives you. Widen only after the repeat purchase holds.
How much stock should the first retailers get?
Less than they will accept. A retailer who sells out and reorders becomes an ally. A retailer sitting on three months of stock blocks your shelf space and blames the product. Small first orders, frequent visits, and a clear policy on unsold stock.
Is influencer marketing useful for a new product?
For a product people can judge from a video, such as food, cosmetics or a gadget, yes, and best with creators in the city where the product is actually available. Sending a product to a creator whose audience cannot buy it produces views and no sales.
When should the price be discounted at launch?
Introductory pricing works when the buyer is choosing between you and a familiar brand at the same price point. It fails when it becomes the price. Put a date on it, tell the trade the date, and hold it.

From the blog

More notes for owners.

Get a free audit of how you sell, and a scored report of where the work is.

Book a free audit