How to Handle Three Times the Volume Without Tripling the Team

The founder asked it plainly. His company handles about thirty quote requests a day. Why not a hundred?

It is a fair question for a small company that sources hard-to-find aerospace parts. The requests arrive by email. Some ask for one part. Some list a thousand part numbers, and nobody is going to work through a thousand lines by hand, one at a time.

In June he put the constraint in a single sentence. He was limited by how many things he could click on.

We ran discovery there this year: eleven interviews, one of them on site, with the findings presented at the end of July. Nothing is built yet. What follows is what we found and what we recommended. There is no result to report.

Where the ceiling is

Every quote request runs the same loop. Someone reads the request and works out exactly what is being asked for. Someone asks suppliers for price and availability. Bids come back by email, each supplier in its own format. Someone chases the suppliers who have not answered. Someone assembles the quote.

Almost every step is a person reading something and typing it somewhere else, or a person waiting. That makes throughput a simple multiplication: the number of people, times how much reading, typing and chasing each of them can do in a day. It is why the honest answer to "why not a hundred" has always been more people.

Quoting is how a sourcing company earns the chance to sell, and in any sourcing business most quotes do not become orders. So much of the effort in any week goes into requests that will not, and every one of them costs the same reading and typing as the ones that will.

What discovery pointed at

Four recommendations, in the order they feed each other.

Read the emails so nobody retypes them. Parse incoming requests and supplier bids into structured data the moment they arrive. A thousand part numbers become a thousand rows without anyone copying them out.

Decide where attention goes. What we called a quote request intelligence engine. Once requests are data, the system can sort them, so the team starts the day on the requests most worth their time rather than whichever arrived first.

Chase suppliers on a schedule. Supplier follow-up that runs itself. The reminder goes when it should, every time, and a person steps in only when a supplier needs a phone call.

Write the founder's judgment down. A great deal of what makes a hard part findable sits in the founder's head: how he approaches a part nobody else can find. We recommended capturing it as a sourcing playbook that the team and the systems can both use.

The order matters. Triage is only as good as the data under it, and the data only exists once the emails are being read. Build the intelligence first and it has nothing to reason about.

Capacity is a design decision

At the readout, their operations manager put it more neatly than I had: build throughput from systems, "rather than by people."

Headcount is the default way to add capacity because it is the only way that needs no design. Hire, train, and the ceiling moves up by one person. It works. It is also slow, and each new person needs the same time with the founder to learn the same judgment, which is the scarcest thing in a company this size.

Capacity built into systems moves the ceiling differently. Parsing does not tire at request number eighty. Follow-up does not forget a supplier on a busy afternoon. The people the company already has spend their day on the parts that need a person: judging an odd request, negotiating with a supplier, talking to a customer. The team still grows, when the business calls for it, not as the fixed price of every extra request.

I am careful with the arithmetic here. At a steady hit rate, more than three times the requests would mean more than three times the orders. Whether the rate holds at higher volume is exactly the thing to measure, because the easiest way to quote a hundred a day is to start quoting worse requests. That is why the triage recommendation matters as much as the parsing. More volume is only useful if it is volume worth having.

Find your own hundred

Most companies have one unit of daily work that sets the ceiling. A quote, an order, a ticket, an application. Find yours, and write down how many you do a day today.

Then walk one through from start to finish with the person who does it. Mark every step where someone reads something and types it somewhere else, and every step where someone waits on someone else. Those steps are where systems carry load. The steps where someone decides are where your people should be spending their day.

And write down your hit rate, or whatever your equivalent is, before anything changes. When the volume goes up, it is the number that tells you whether the extra volume was worth having.