In this article
Across ⟨N⟩ completed orders on VendorStocks between ⟨start month⟩ and ⟨end month⟩, the median time from RFQ to delivery was ⟨X⟩ days. ⟨Stage name⟩ accounted for ⟨Y⟩ of those days — more than any other stage, and more than the physical movement of goods. The bottleneck in Indian electrical procurement is not logistics. It is ⟨stage⟩.
Nobody in this trade publishes this kind of measurement, which is the main reason we are doing it. What follows includes the parts that did not flatter us.
What we measured, and what we left out
Sample. ⟨N⟩ orders that ran the full sequence — RFQ raised, quotation issued, purchase order released, invoice raised, payment confirmed, delivery reported — between ⟨dates⟩.
Excluded. Orders cancelled before PO. Orders where Skylights Energy was itself the buyer or the seller, because we can move faster on our own transactions and including them would flatter the numbers. Orders with incomplete timestamps.
Timestamps. Taken from platform records at six points: RFQ creation, first quotation received, purchase order issued, proforma issued, tax invoice raised, payment confirmed.
The weak figure. Delivery confirmation is seller-reported. We do not have a carrier feed and we are not going to pretend we do, so the final stage of every cycle carries whatever optimism the seller had on the day he marked it delivered. Treat that stage as the softest number in the table and everything before it as solid.
Stage by stage
| Stage | Median days | 90th percentile | n |
|---|---|---|---|
| RFQ raised → first quotation | ⟨ ⟩ | ⟨ ⟩ | ⟨ ⟩ |
| First quotation → PO released | ⟨ ⟩ | ⟨ ⟩ | ⟨ ⟩ |
| PO → proforma issued | ⟨ ⟩ | ⟨ ⟩ | ⟨ ⟩ |
| Proforma → payment confirmed | ⟨ ⟩ | ⟨ ⟩ | ⟨ ⟩ |
| Payment → dispatch | ⟨ ⟩ | ⟨ ⟩ | ⟨ ⟩ |
| Dispatch → delivery reported | ⟨ ⟩ | ⟨ ⟩ | ⟨ ⟩ |
| RFQ → delivery | ⟨ ⟩ | ⟨ ⟩ | ⟨ ⟩ |
Two things to notice when you fill this in. First, look at the gap between the median and the 90th percentile in each row — a stage with a tight spread is a process, and a stage with a wide spread is a queue with a human in it. Second, the stages that involve a decision are almost always slower than the stages that involve moving goods, and that is the finding that matters.
The stage that surprised us
⟨Write this section after you see the data. The honest version: name the stage, say what you expected, say what it actually was, and resist the urge to explain it away. If the slow stage is one we control, say that too — a data post where the platform comes out blameless reads as marketing and will be treated as marketing.⟩
Where the spread is widest
Medians hide the orders that go wrong, and the orders that go wrong are the ones a buyer remembers.
⟨Identify which stage has the widest gap between median and 90th percentile, then describe what an outlier order looks like. Typical candidates in this trade: orders waiting on an internal approval that nobody has chased, orders where the buyer’s payment cleared into a long weekend, orders for an item the seller had to indent rather than hold.⟩
The useful question for a buyer is not what the average order takes. It is what the slow ten per cent look like and whether your order resembles them.
What a buyer can compress, and what he can’t
Some of this cycle is yours to fix and some of it is not.
You control how fast you release the PO after the quotation arrives, how quickly your payment clears, and whether your RFQ was complete enough to be quoted against without a follow-up. In most procurement teams the biggest single saving available is not a better supplier — it is an internal approval that currently sits with one person who is travelling.
You do not control the seller’s dispatch cycle, the transporter, or whether the item is in stock at the branch you happen to be talking to. What you can do is ask for a dispatch date in writing before you release the PO, which converts an uncertainty into a commitment you can plan around.
Nobody controls a long weekend, a festival, or the last three days of a financial quarter, and any lead-time promise made without reference to the calendar is worth less than it sounds.
What we got wrong in this measurement
Three things, and we would rather say them here than have somebody find them.
Delivery dates are seller-reported, as noted above. That makes the final stage softer than the rest of the table.
The sample skews toward the brands we carry most, which means this is more a picture of terminal blocks and LT switchgear than of Indian electrical trade generally. A cycle for a bought-out HT item would look nothing like this.
And ⟨N⟩ orders from one platform is a keyhole view of a very large market. We are publishing it because nobody else publishes anything, not because we think it settles the question.
We will run the same measurement again next quarter with the same method, and you will be able to see whether these numbers hold or whether we were describing a season.
Why this matters more than a lead-time promise
Every supplier in this industry quotes a lead time. Almost none of them measure the full cycle, because the full cycle includes the buyer’s own delays and no supplier is going to raise that.
The gap between “we dispatch in three days” and “your order takes ⟨X⟩ days from the moment you asked for a price” is where procurement planning actually lives. If you are scheduling a panel build off a supplier’s dispatch promise rather than off your own historical cycle time, you are planning against the wrong number, and you have probably noticed that without being able to say why.
Measure your own. Six timestamps, fifty orders, one afternoon in a spreadsheet. It is the most useful thing a purchase department can do for itself this quarter, and you do not need a platform to do it.
Related reading: “Ex-stock” doesn’t mean in stock · The real cost of running procurement on WhatsApp · Why dealers lose deals after the quotation
All insights