Three clients, three different worries

Each of these three came from a video. The clients sat down and talked, at length, about what the work had been like. What is printed below is a summary of what each of them said. It keeps the substance and the specifics. It is not a word-for-word transcript.

That distinction matters enough to put at the top rather than in a footnote. A quote that has been tidied up and then presented as somebody's exact words is a small dishonesty. It is also the kind that makes a testimonial worth very little. The situations, the numbers and the judgements are all theirs. The sentence construction is ours. Where a summary and the original wording differ, the original wording counts. These will be replaced with the filmed wording once each transcript is reconciled.

The attributions are not summaries. Every name, role, company and engagement window below is exact.

This is the worry that survives every reference call. It cannot be answered by anything a company says about itself. The answer usually arrives somewhere around the second year.

The second sentence is the checkable one. Responding before the client's own operations team noticed is a behaviour with a timestamp on it. An events platform finds that out on a bad night.

The usual trade is real. A team quoting a short timeline is often quoting on a problem it has not looked at yet. What this one turns on is where the time went. It went into the first call.

The last clause is worth reading twice. It does not flatter us in the way a testimonial usually does. He is saying the one defect traced back to his own specification. He was close enough to the work to know where the fault was. That is a better claim than a client who never found out.

Everybody who has commissioned software has learned to add a margin to whatever date they are given. The claim below is about when bad news arrives, rather than about speed.

One thing to be straight about: that is a summary of two separate testimonials, from two founders, presented together. They are printed as one because that is how the substance was recorded. Guessing which sentence came from which person would tidy the layout and falsify the attribution. Splitting them would mean inventing an attribution. An awkward honest one is worth more.

A testimonial is somebody else's evidence about somebody else's project. The way to make it useful is to turn it back into a question. Then ask that question of everyone you are talking to, in the same words.

Ask who answers at two in the morning. Ask what happened the last time something broke at that hour. A team that has been through it tells you a story with a date in it. Listen for whether you get that or a description of a process.

Ask what happens in the scoping call, and what you will be holding when it ends. The integration above shipped in three weeks. The first hour went into the specification rather than the estimate.

Ask how far in advance you hear that a date has moved, and who tells you. Ask us, and ask whoever else is bidding. Compare the answers rather than the timelines.

Those three questions are everything above, in a usable form. They work on a call that has nothing to do with us.

Three is three. It does not make a sample, a satisfaction rate, or evidence about how the other engagements went. We will not build an average out of it or attach a star rating. Neither would mean anything.

They also do not cover every kind of work. They cover a live events platform, a vehicle rental operator, and two founders running work across several partners. Your own situation may sit a long way from all three. If it does, we have not shown you anybody like you, and that is worth weighing.

Each of them is a moment rather than a verdict. Each one describes a particular stretch of a particular engagement. The Eventerprise engagement ran from 2017 to 2020. The quote captures one point inside that window.

If you want the work itself rather than what clients said about it, the case studies are the better read. They are anonymous by policy, even where the testimonial is named.

What a wrong choice of software engineering partner costs is two quarters spent finding out. Add the discarded version of the product, and the internal credibility of whoever signed. The invoice is the small part of it. Gary's testimonial above starts with two teams that had already missed it. That is the ordinary version of the cost. It is also why a third attempt gets scrutinised in a way the first never was.

That is the whole reason a stranger's opinion carries any weight here. A stranger who has already spent those two quarters knows which questions are load-bearing.