Lean Six Sigma in sales

The commonest finding in a sales process is response time. A lead arrives and waits: hours in the good cases, days in the ordinary ones, while the buyer, who contacted three suppliers on the same afternoon, is being answered by somebody else. No amount of skill at the meeting recovers a lead that went cold before the meeting was offered.

The second is the quote. In most organisations a quote crosses two or three functions, waits for a price approval, and comes back with an error rate nobody measures. Quote turnaround and quote accuracy are ordinary process metrics, and they move the win rate more reliably than any training course.

The third is that the pipeline is measured as a total rather than as a set of stages. A pipeline value of two million tells you nothing. Conversion from stage to stage, and the time work spends in each, tells you where deals actually die, and it is usually one specific transition that everyone had a feeling about and nobody had counted.

A sales project starts by drawing the pipeline as a value stream, not a funnel diagram. A funnel is a picture of hope. A value stream shows each step, who does it, how long the work takes and how long it waits, and the waiting is where the losses are.

Measure means taking the numbers out of the system rather than out of the sales meeting, and it usually exposes that the system is not being kept properly. That is itself a finding: a pipeline that is not maintained cannot be managed, and cleaning it is the first improvement rather than a prerequisite for one.

Analyse needs care with small numbers. A team closing twelve deals a quarter cannot tell a real change in conversion from ordinary variation with a month of data, and treating noise as signal is how sales processes get changed every quarter and improve never. When samples are small, the honest tools are the ones that carry their uncertainty with them.

Improve is usually unglamorous: route leads automatically, pre approve a price band so quotes stop waiting, agree what qualified actually means so two people mean the same thing. Control is a weekly chart of response time and stage conversion, owned by the sales manager rather than by an analyst.

The relationship is not a process step. A good seller reads a room, chooses when to push and when to wait, and knows which questions land with which buyer. Standardising that produces a script, and a script produces the flat, distrusted call everyone recognises. Improve the machinery around the conversation and leave the conversation to the person having it.

Small samples are the real technical difficulty. Sales data is thin, seasonal and full of exceptions, and most of the apparent effects in a quarter are noise. Any sales improvement programme that declares victory on one good month has learned nothing, and will unlearn it next month.

The last risk is measurement that changes behaviour for the worse. Measure calls made and you get calls made. Measure meetings booked and you get unqualified meetings. Every sales metric needs its counter measure beside it, or the number improves while the business does not.