Retail Conversion Rate Benchmarks in India by Category

How to calculate conversion, and which one you are calculating

1. Choose the denominator and write it down. Door count from a counter, walk-ins from a register, sales visits after service visits are removed, enquiries logged in the CRM, or consented conversations captured. Each gives a different rate; none is wrong, and mixing them across stores is the commonest benchmarking error.
2. Choose the numerator. Bills is the usual one. Buying groups is better where families shop together, because one family is one visitor group and one bill. Bookings, not retail, in automobile; enrolments, not enquiries, in education.
3. Choose the window. Same day is the floor's number. Seven, fourteen or thirty days including recovered sales is the store's number, and the gap between the two is the value of the follow-up habit.
4. Divide, per store, per week. Conversion is bills divided by visitors, as a percentage. Do it weekly because salary dates, weekends and festivals move it too much for a daily read and too little for a monthly one.
5. Report the denominator's coverage next to the rate. A conversation-based rate on a week where advisors captured a third of walk-ins is a guess about a third of the store. The footfall-and-conversation approach is at /learn/walk-in-conversion-tracking-retail/.

The footfall denominator problem

The same durables store, on the same Saturday, can report 30 percent, 42 percent or 55 percent conversion depending on how visitors were counted. A door counter counted 300 entries, including staff, delivery pickups, a family counted as four and the man who stepped in to take a call. The register counted 210 walk-ins because the greeter missed the crowd at 6 pm. Consented conversations captured 165, because the unattended were never recorded. Ninety bills were raised. None of the three rates is dishonest; each measures something different.

This is why published benchmarks, where they exist, rarely say what they divided by, and why a national number in a vendor's brochure cannot be compared with a store's own rate. It is also why the useful comparison is between stores counted the same way, which usually means the retailer's own stores.

• Counter over a door that also serves a service desk, a cafe or a pickup point: inflates the denominator, deflates conversion. Common in telecom and in mall durables stores.
• Groups: a family of four counted as four entries and one bill reads as 25 percent when it is 100. Group-aware counting or a register that logs parties fixes it.
• Staff and repeat entries: an advisor stepping out for a smoke is two entries. Most counters can exclude a staff lane; few stores set it up.
• Register at peak: the human count fails precisely when it matters. The gap between counter and register on a Saturday evening is a fair estimate of unattended walk-ins, and that is a useful number in its own right.
• Conversation capture: measures the floor's performance on the customers it spoke to, and nothing about the ones it did not. Report it with coverage, always.

Working ranges by category

These are working ranges from assisted retail practice, not published statistics. They are for walk-in to same-day purchase on a walk-in register or a group-aware counter, with an extended-window figure where the category's purchase takes more than one visit. Treat the low end as a store that is losing at attendance or the close, the high end as a store with strong intent footfall, and anything outside the range as a denominator question before it is a performance question.

What public sources do and do not publish

Anyone searching for an Indian retail conversion benchmark finds vendor pages quoting a global figure with no denominator and no country. The Indian public sources that exist report adjacent numbers, and it is worth knowing which so the gap is not mistaken for ignorance.

• The Federation of Automobile Dealers Associations (FADA) publishes monthly retail registration counts by vehicle segment from VAHAN data. It does not publish showroom walk-ins or enquiry-to-retail conversion; those stay inside dealer management systems and OEM reviews.
• The Retailers Association of India (RAI) publishes a periodic Retail Business Survey reporting sales growth by category and region against the previous year. It does not publish conversion or footfall-to-sales ratios.
• The Telecom Regulatory Authority of India (TRAI) publishes monthly subscriber additions and mobile number portability requests by operator and circle. Nothing in it describes store-level conversion.
• Listed retailers and jewellers occasionally mention walk-in growth or footfall trends in investor presentations and earnings calls. Conversion is rarely disclosed, and where a figure appears the denominator is not defined.
• Footfall-counter and retail-analytics vendors publish conversion figures, mostly for Western speciality retail, mostly without a stated denominator. They are useful for the shape of the argument, not as an Indian benchmark.
• Everything in the table above is therefore a working range from assisted retail practice, and should be quoted as such.

Why store-to-store variance matters more than the national number

Suppose a durables chain of forty stores reads a national working range of 20 to 40 percent and its own average of 31 percent, and concludes it is fine. Inside that average, on the same counting method, its stores run from 18 to 44 percent. The store at 18 percent shares a city, a range and a scheme with the store at 44. The difference between them is attendance, demonstration rate, the EMI quote and the number-taken habit, all of which the chain controls, and the gap between the chain's worst and best store is worth more than the gap between the chain's average and any published number.

The variance is also where the coaching material is. The 44 percent store has an answer to the online price objection that the 18 percent store does not. Finding it and moving it is a matter of weeks, and the method for ranking stores within a footfall band and finding the band's best rate at each stage is at /learn/how-to-compare-retail-store-performance/ and /learn/how-to-identify-underperforming-retail-stores/.

A national benchmark, even a good one, answers a question a category head rarely needs to ask. Whether the chain is at 31 or 34 percent against India is less useful than whether the bottom ten stores can reach the band's median, which is a concrete number of sales per week per store, and a plan.

Using the ranges honestly

1. Count every store the same way before comparing any two. If some stores have counters and some registers, compare within each group or convert with a measured ratio for each store.
2. Group stores by footfall band and format: metro mall, high street, tier-two standalone. A 25 percent metro mall store and a 25 percent tier-two store are not the same store.
3. Rank within the band on same-day and on the extended window, and read the gap. A store strong on same-day and weak on the window has no follow-up habit; the reverse has a floor problem the follow-up is hiding.
4. Use the working range only as a sanity check on the counting. A store reporting 70 percent in durables is counting sales visits or bills per group, not walk-ins. A store at 8 percent has a counter over the wrong door or an attendance problem, and the one-week diagnosis is at /learn/why-store-sales-are-down-when-footfall-is-not/.
5. Read conversion next to the numbers underneath it: attended, demonstrated, objection answered, number taken. Two stores at 30 percent with different shapes need different fixes. Borentis reports those stages from consented conversations in Hindi, English and Hinglish, with coverage shown next to every rate, which is one way to make the comparison honest; a register and a numbers log are another.