The 4 numbers every Indian warehouse manager should know on Monday morning

When I started running warehouses, I built dashboards with 40 KPIs. Pick rate. Pack rate. Order cycle time. Inventory turnover. On-time dispatch. Dock-to-stock. Cross-dock rate. Each one felt important. Each one had a chart, a target, a colour-coded red/yellow/green. None of them helped me actually run the floor better.

After about a year of drowning in data, I threw the dashboard away. I picked four numbers. I made them the only things the warehouse team and I talked about. And — to my surprise — the operation got noticeably better. Because when you only have four things to focus on, you actually focus on them.

Here are the four. Each is one line on a spreadsheet. Each takes less than 5 minutes to calculate on Monday morning. Each is directly tied to P&L.

1. Pick accuracy (target: 99.5%+)

What it is: the percentage of picked orders that ship the right items. Calculated as (orders shipped without a return due to picking error) / (total orders shipped).

Why it matters: a pick error is the most expensive thing that can happen in a warehouse. The order ships, the customer gets the wrong item, you pay reverse shipping, you pay re-shipping, you take a customer service hit, and you eat a RTO charge. Conservatively, each pick error costs ₹400-600 in cash, plus the lifetime-value hit.

Indian context: most Indian warehouses run at 96-98% pick accuracy. Sounds fine. On 1,000 orders a day, that's 20-40 wrong shipments. At ₹500/each, that's ₹10-20K/day, or ₹3-7L/month. Most founders I talk to don't realise this is happening because the errors are hidden in the returns data.

How to improve it: wave-based picking with barcode verification at the bin (not at packing). The picker scans the bin, the system tells them which SKU is expected. If they scan a different SKU, the system refuses. This alone takes pick accuracy from 98% to 99.5%+ within a month.

2. RTO rate (target: <10% for COD, <3% for prepaid)

What it is: the percentage of orders that ship and come back undelivered. Includes customer refusals, NDR (non-delivery reports), and address issues.

Why it matters: RTO is the single biggest cost centre in Indian ecommerce. The forward shipping is paid but the return shipping is also paid. The product is often unsellable. The customer experience is destroyed. COD RTO rates are 12-20% across the industry; prepaid RTO rates are 2-5%.

Indian context: the highest RTO pincodes in India are well-known (most Tier-2/3 cities in UP, Bihar, NE). Smart brands block or pre-verify these pincodes for COD. Smart-er brands also pre-verify all COD orders via IVR before shipping.

How to improve it: split your metrics by pincode. Block the worst 50 pincodes for COD. Pre-verify the next 200 via IVR. Auto-cancel orders that fail verification. This single change usually drops RTO from 15% to 9% on COD, saving ₹5-10L/month at ₹1 Cr GMV.

3. Dispatch TAT (target: <4 hours for same-day, <24 hours for next-day)

What it is: the time from when an order lands in your system to when it ships out the door. Calculated as p95(order_shipped_at - order_created_at). Use p95, not average — the average hides the worst 20%.

Why it matters: in Indian ecommerce, dispatch TAT is the single biggest driver of customer satisfaction. Amazon's 24-hour TAT is the industry benchmark. Nykaa and Myntra penalise you for every hour of delay. Your customer never forgets the order that took 4 days to ship "because it was processing."

Indian context: most brands I've worked with have a 12-18 hour p95 TAT. They think they're fast because their average is 4 hours. But 20% of their orders are taking 18+ hours, and those are the orders that get cancelled, the ones that get bad reviews, the ones whose customers never come back.

How to improve it: cut the order-to-picklist step. If you're still printing picklists at 11am and dispatching at 5pm, you have a 6-hour latency that software can cut to 30 minutes. Wave-based picking with auto-generated picklists on order arrival is the unlock.

4. Dead stock % (target: <5% of total inventory value)

What it is: the percentage of your inventory value that hasn't shipped in the last 90 days. Calculated as (value of SKUs with no outbound movement in 90d) / (total inventory value).

Why it matters: dead stock is the silent killer of D2C working capital. Every brand I've audited has 8-15% of their capital tied up in SKUs that will never sell at full price. Some will markdown. Some will become return gifts. Some will be written off entirely. All of it is money that could be in new SKUs that actually move.

Indian context: the typical causes of dead stock in India are: (1) a forecast for a festive season that didn't materialise, (2) a SKU launched with too much initial buy, (3) a shade/size variant that nobody wants (the lipstick in "warm nude" that the founder's wife likes but nobody else does), (4) packaging change that left the old stock orphaned.

How to improve it: look at this number every Monday. The 90-day threshold is a starting point; some categories (like fashion) should use 60-day. The first time you see the number, you'll be shocked. The second time, you'll start making smaller initial buys on new SKUs. The third time, you'll add a clearance workflow to the WMS. The fourth time, you'll be at 5%.

What about the other 36 KPIs?

They're not useless. They're useful for diagnosing specific problems. If your pick accuracy drops one week, you pull the pick-rate-per-picker report and find out which picker is making errors. If your dispatch TAT spikes, you pull the order-status-time report and find out which step is slow.

But as Monday-morning numbers — the numbers that define "is the warehouse healthy" — these four are it. Print them on a piece of paper. Tape it next to the time clock. Ask the floor supervisor about them every Monday at 9am. Within a quarter, you'll have the best-run warehouse in your category.

If you want to see what these four numbers look like in a WMS dashboard, see the Analytics feature. And if you want to see your own four numbers, start a 14-day trial — they show up on day one.