Overstocked on the wrong things. Out of stock on the right ones.
It costs retailers 6.2% of sales every year — more than many of them make in net profit. Plavista is planning software for the retailers and consumer brands the enterprise platforms were never built for. Runs on the ERP you already have.
Worldwide out-of-stock rate. It has barely moved in twenty years.
Gruen, Corsten & Bharadwaj, GMAOf out-of-stocks come from the retailer's own ordering and forecasting — not from suppliers.
Kaizen InstituteEmpty-shelf encounters where the sale is gone: 31% buy it elsewhere, 9% don't buy at all.
GMA worldwide studySales uplift from correcting inventory-record accuracy alone, even off a strong baseline.
ECR Retail LossNone of this is bad luck, and most of it isn't the supplier. It's forty thousand ordering decisions a week, made by people who don't have the time or the tools to think about them.
“I helped build one of the enterprise planning platforms, then spent fifteen years implementing the others. They're excellent — and they were never designed for you.”
Ranga Mallipudi, founder — twenty years in retail planning, on both sides of the softwareA year, for forecasting and replenishment alone, at a 300-store grocer. Two modules out of a suite.
VendorBenchmark, 2026Implementation, on top of the first-year subscription. The partner bill is usually the same order as the licence.
VendorBenchmark, 2026A typical enterprise deployment, before any of it reaches a planner's screen.
Published implementation timelinesSo a ₹2,000 crore retailer gets the quote, does the arithmetic, and goes back to spreadsheets. Given those numbers, that is usually the right decision — and it is exactly why this exists.
Plavista Fresh: four things, every night.
Decides what to order
Every fresh item in every store, every day, plus the collection-centre buy — with delivery days, shelf life, crate sizes and display minimums already accounted for. And it shows its reasoning on any line you question.
Hands your team a short list
The few stores and lines that genuinely need a human today: stores wasting or running out far more than similar stores, stock records that don't add up, packs too big to sell before expiry — each with the reason. Everything else is proposed with its reasons, approved in a tap, logged and reversible.
Catches what your reports hide
Demand that never reached your sales report because the shelf was empty, estimated from when it ran out. Stock records that drift from the shelf, with a short daily count of the items most likely to be wrong.
Proves what it was worth
Waste at cost, lost sales and availability, measured against comparable stores still ordering the old way. Your finance team signs the number. We don't.
Four commitments.
We sit on top of your ERP
We read its standard export and return a file it can take in: orders for a retailer, dispatch quantities for a brand. We never become your system of record, and nothing you have already invested in gets thrown away.
Proof before you rely on it
First a backtest on your own history, against the reorder rule you run today. Then a few weeks side by side in a few stores, against matched stores that keep ordering as before, with a stop rule agreed in advance. You see exactly what it would have done before you trust it with a rupee.
Your policy, your dials
How much availability, how much waste, category by category — these are commercial decisions, so your category managers set them on screen. Not parameters buried in a config file by a consultant.
Outcome-linked fees for early partners
For our first partners, part of the fee can be a share of what we actually save you, measured against control stores locked before go-live and agreed with your finance team. If it doesn't work, we don't earn that part.
One platform, built in the order that pays for itself.
We'd rather be honest about what exists today than show you a brochure. The products share a common planning foundation: one way of reading store, item and day data, of treating sold-out days as hidden demand, of valuing waste at cost and of explaining every number. On top of it, each product has its own forecasting and decision models, because ordering tomorrow's fresh and planning a launch are different decisions. Built in this order.
Designed to read standard exports from the ERP you already run, or a nightly file drop if that's what you have. Plavista is not affiliated with any ERP vendor.
The diagnostic. Problem first, product second.
We find where the money actually goes in your business — across stock, availability, returns, expiry and channel — before selling anyone software. Retailers start with the diagnostic; food brands start with the launch review.
Twelve months of sales, stock, purchase order and write-off data. Under NDA.
Please don't clean it. Send it exactly as it comes out of your system — the mess is usually where the findings are.
A written analysis of where the money went, broken down by cause, with the top ten fixes ranked by what each is worth, and a backtest against the reorder rule you run today.
Yours to keep and act on, whatever happens next.
A fixed fee, agreed before we start — and credited in full against your first year if you decide to go ahead.
About six weeks from the day the data lands. No obligation at the end of it.
I began in the R&D team of one of the major enterprise planning platforms, helping build its merchandise planning suite — merchandise financial planning, assortment planning and size profile optimisation — largely from the ground up. Retailers around the world still run it today.
Then fifteen years on the other side of the table, leading implementations of that platform and its main competitors for grocery, hypermarket and fashion retailers across the US, UK, Middle East, South Africa and India, through global consulting firms.
Those platforms are very good, and they are built for a scale of retailer most of the industry will never reach. Everyone else is left with spreadsheets and instinct. That's the gap Plavista is being built into.
Alongside the platform, I advise retailers and consumer brands on planning transformation, platform selection and operating model design, whichever platform they run. About advisory
Things you can check in your own data this week.
The history you haven't written yet
A new retailer's demand history has not been written yet. What to instrument before the first store opens, and what it changes about platform selection.
The blind window
Fresh waste is committed days before anyone can see it. The gap between the moment an order becomes irreversible and the moment demand becomes knowable is where the money goes.
Your sales report is lying to you
Sales recorded during a stockout aren't demand. How forecasts learn from shortages they caused, and why the availability report never notices.
The system says eight. The shelf has none.
Phantom inventory quietly switches off reordering. Why it happens, why almost no software catches it, and how to find it in your own data.