A self-scored diagnostic for engineering leaders
You are one signature away from spending a year on the wrong problem.
Three different problems make delivery look slow. They need opposite fixes, and the plan on your desk is the right answer to exactly one of them. By Friday you can know which one it is, out of your own history.
- What you end up with which of the three you have, and what to change first
- What it takes an afternoon, on your own
- What you have to buy nothing
Where should I send the diagnostic?
You get the Delivery Diagnostic straight away. After that I email you twelve short notes, one a day: why delivery stalls, and what to change. They come from me, Alex Fedorov, at 10x Software Engineering. One click stops them at any time, and the diagnostic stays yours either way.
I store your email address to send you the diagnostic and those notes, and nothing else. Privacy policy.
Awaiting signature
Engineering plan, next financial year
- Investment
- Term
- Expected outcome
Approved Date
The complaint is the same. The cause is not.
From the outside, delivery is slow. That is all anyone above you can see. Underneath it there are three different problems, they produce the same complaint, and they need opposite fixes.
Your dashboard will not separate them. It reports on the part of the process engineering already controls, which is why it can look healthy in the same week the business is telling your CEO it is waiting on you. So when you cannot tell the three apart, you do what everybody does. You pick one of these and find out in about a year.
You hire
The easiest thing to get signed off and the hardest to undo. Six months to fill the roles. Six more before the new people are worth what they cost. At the end of it you are running a larger organisation with more boundaries inside it than the one you started with.
Spent: a year, a permanent line on the budget, and a department that is now harder to change.
You run a transformation
New process, new ceremonies, a framework or a consultancy, everybody's week rearranged. It takes eighteen months to find out whether it worked. By then so much else has changed that nobody can say which part was the cause, including you.
Spent: eighteen months, and the team's willingness to believe you the next time you ask them to change how they work.
You buy the tooling
Licences for everyone, a rollout, and developers who genuinely produce code faster. That part is real. Then nothing arrives any sooner, the business notices no difference, and somebody starts asking what the spend was for.
Spent: the budget, and the argument that engineering investment pays back, which you will need again next year.
Each of these can be the right move. None of them is a way of finding out.
What arrives
This is a way of finding out.
Eleven pages you work through rather than read. You fill it in as you go, and by the end of the afternoon the answer is written down in front of you.
- Format PDF, eleven pages
- Runs on your own delivery history
- Costs your email address
- Yours to keep, and to pass on
An afternoon against a year
You cannot fix a problem you have not identified. You also cannot spend a year identifying it. So this had to be cheap enough to run before you commit to anything.
What you hold at the end of it: which of the three problems you have, how much weight that answer deserves, and the first thing to change. Specific to your organisation, out of your own history, and not a band off somebody's maturity model.
What you do not need: new tooling, a vendor, an instrumentation project, a consultant in the building, or anybody's permission. You can do the whole thing yourself with what your organisation already has.
And if it turns out I am describing somebody else's company, you have lost an afternoon and you know one true thing about your delivery that you did not know on Monday.
- 3problems it tells apart
- 1afternoon to run it
- 0tools to install
I cannot prove any of this to you. You can prove it to yourself before Friday.
Who is writing this
I am Alex Fedorov. I have built and run engineering organisations, and I wrote 10x Software Engineering Delivery. This is the problem I work on.
Most failed delivery work I have seen was a correct intervention aimed at the wrong problem. Competent people, real effort, real money, and the same complaint at the end of it. Nobody in those rooms was lazy or stupid. They were confident about which problem they had, and they were wrong, and there was nothing cheap in the building that would have told them so.
This is not built out of my results. It is built to run against yours, this week, and to give you an answer you can act on the same day.