BigBasket ads: what you can buy and how to judge it

If you sell atta, shampoo, coffee or nappies in India, someone has told you to put money into BigBasket.

This page is about whether that is worth doing, what is actually buyable, and how to grade it honestly. It assumes you already know the category — if you do not, start with our explainer on how retail media networks work and come back. We will not repeat it here.

Where BigBasket stands in October 2026

You cannot plan this spend without knowing what the platform has become. BigBasket is not the weekly-slot delivery service it was in 2019.

Two honest caveats on that table. Dark-store counts move monthly and reports disagree — trade coverage in July 2026 put BigBasket at more than 700 dark stores, below the company's own December 2025 number. Treat any single count as a snapshot.

Second, the business is lossmaking. BigBasket accounted for roughly 64% of parent Tata Digital's ₹4,974 crore FY26 loss, per Inc42's reading of the Tata Sons FY26 annual report. Advertising is one of the few lines growing faster than the top line. That tells you the commercial pressure on the ad business is upward — expect more inventory and more selling, not less.

The cycle, not the session.

What is buyable on BigBasket, and is it self-serve?

Here is the part most articles get wrong, so we will be precise about what we could and could not verify.

BigBasket does not publish an advertising page for brands. We looked for one on 19 October 2026 and found no official page listing formats, specifications or rate structures, and no public self-serve console of the kind Amazon or Flipkart run. If one exists, it is behind a partner login rather than open signup.

What we can say with evidence:

• Buying is sales-led. Access comes through BigBasket's brand or category team, usually attached to your supply relationship, or through a media agency that already holds the relationship.
• There is an in-house ads team. BigBasket has been hiring for ad optimisation and brand intelligence roles whose brief includes aligning campaign execution with "advertiser expectations" and growing advertising revenue — a managed-service shape, not a self-serve one. That listing closed in September 2026.
• Resellers list display-style formats. Third-party media buying site The Media Ant lists four BigBasket formats — banner, video, roadblock and scratch card — and when we checked on 19 October 2026 it marked all four as having no inventory available. The page carries no publication date, so treat it as indicative only.
• Sponsored listings and search placements are the formats brands most often describe using, but we found no dated primary source confirming their current specifications. We are not going to invent them.

No CPC, CPM or ROAS benchmark for BigBasket is published by the platform. Any number you see quoted in a blog post is either an agency's own account average or made up. We have left them out.

Why grocery advertising is judged differently

This is the real argument of the page, and the reason BigBasket should not be planned like a fashion marketplace.

Two things are true about groceries that are not true of most categories:

1. The purchase repeats on a cycle. Toothpaste runs out. A sofa does not.
2. The purchase happens inside a planned basket. The shopper is filling a list of 15–30 items, not evaluating one product. BigBasket's average order value was stated at ₹650 by its chief buying and merchandising officer in December 2025 — that is a shop, not an impulse grab.

That second point separates BigBasket from impulse-led quick commerce. The relevant contrast platform is Instamart, which we cover in our guide to advertising on Swiggy, and the planned-basket pattern is also why BigBasket behaves differently from a general marketplace like Flipkart.

So the win condition is not "a sale today". It is basket inclusion — becoming a default line on a repeating list. That changes three decisions:

Working through the maths

Clearly hypothetical numbers — put your own in.

Buyers acquired via BigBasket ads     1,000
Media cost per acquired buyer          Rs 120   -> Rs 1,20,000 spent
Gross profit on first order            Rs 55    -> Rs 55,000 earned

Looks like a failure. Now add the repeat cycle:
Repeat rate within 90 days             38%      -> 380 buyers
Average repeat orders each             2.4      -> 912 orders
Gross profit on those                  Rs 55    -> Rs 50,160 earned

Total 90-day gross profit              Rs 1,05,160  vs  Rs 1,20,000 spent

Still short — but look at what decided it. Not the creative, not the bid. The repeat rate. At roughly 49% instead of 38%, the same campaign at the same cost per buyer clears its media cost inside 90 days.

That is the whole case for measuring grocery ads on retention. If you want the formal version of this calculation, see customer lifetime value, and the mirror-image metric in churn rate.

Availability comes before advertising

An ad on a grocery platform sells a specific pack, in a specific dark store, on a specific day. If the pack is not in that dark store, the ad either does not serve or sends the shopper to a substitute.

So before you brief a campaign:

• Check serviceability by pincode for your top SKUs, not just that the listing exists.
• Fix the listing itself — correct pack size, weight, images and a title a shopper would actually type.
• Watch stock on the SKUs you promote. Spend aimed at an out-of-stock pack is dead money, and you usually find out late.
• Know that you compete with the landlord. Private label is 36–37% of BigBasket's turnover and heading for ~40%. In many categories the shelf's best-placed rival is the platform's own brand.

That last point is structural, not a quirk. It is the same tension D2C brands manage everywhere, and it is one reason to keep an owned channel strong alongside platform spend — the argument we make in our D2C marketing strategy guide.

What the platform reports, and what it does not

Reporting on BigBasket arrives from BigBasket. There is no independent meter.

The honest answer to "did BigBasket spend grow my business" comes from outside the dashboard. Hold a region or a SKU back, run it dark, and compare. That is incrementality testing, and it is the only method that survives the platform marking its own homework. The wider problem of stitching platform-reported numbers into one view is covered in marketing attribution.

How we judge grocery-platform spend (DH's own view)

This is our working rule set, not an industry standard. We apply it when a client's attribution sits inside a platform we do not control.

1. We discount platform-reported ROAS by default and treat it as a directional ranking tool between our own campaigns, nothing more.
2. We agree one number before launch. For grocery that is revenue per acquired buyer at day 90, taken from the brand's own data where a first-party identity exists. If the brand cannot see repeat behaviour at all, we say so out loud before spending, because then nobody can grade the channel.
3. We run a geographic holdout wherever the platform allows city-level control. Imperfect beats unmeasured.
4. We cap the test at an amount the brand can afford to write off, and we write the write-off scenario into the plan. A sales-led channel with no public rate card deserves that caution.
5. We check availability weekly during a flight. In our experience, more grocery-platform campaigns underperform because of stock and listing faults than because of bidding or creative.

None of this is exotic. It is ordinary media buying discipline applied to a channel that reports on itself.

Who should bother, and who should not

If three or more rows land on the right, put the budget somewhere you can measure. Our performance marketing guide covers how to sequence channels when one of them is opaque, and ecommerce marketing covers the all-channel split.

What a disciplined BigBasket test gives you

Run properly, this spend buys something more durable than a week of orders. It buys a line on a repeating shopping list, and a clear number telling you what a grocery buyer is worth after 90 days.

That number is portable. It prices every other channel you test next year, including the ones that report honestly.