Short answer
#What this calculates, and what it deliberately leaves out
Two costs, because they are the two that are quantifiable without inventing anything:
- Reconstruction time. Engineers being paid to find out what the previous team already knew. This is the cost that hides, because it appears in a budget as ordinary salary rather than as the price of a transition.
- Delayed revenue. Only if something is genuinely blocked. Left at zero by default, because the honest answer for many companies is that nothing is.
Several real costs are left out on purpose, because putting a number on them would require inventing one:
Not counted here
- Opportunity cost — the work those engineers would otherwise have done.
- Mistakes made from an incomplete picture, and the cost of undoing them.
- Incidents that happen because nobody understood a subsystem yet.
- The management and coordination time a transition consumes.
- Customer impact from slower delivery during the period.
- Anything that turns out to be irrecoverable, which is a different kind of cost and not one you can average.
#Why there is no 1ADK figure on this page
The claim this page does not make
You will not find a figure here attributing a saving to 1ADK, and there is not one anywhere on this site. Nobody can know what your handover would have cost without a particular tool, because the counterfactual does not exist. Every vendor calculator that produces a saving is producing a number about a world it cannot observe.
What is defensible is narrower: the reconstruction weeks are the largest and most compressible line above, and deriving a technical picture from the system takes hours rather than weeks. What that is worth in your case is your arithmetic to do, with your figures, on this page.
#Getting a defensible number of weeks
The weeks figure dominates the result and is the one people guess at. Three ways to make it better than a guess, in increasing order of effort:
-
Ask two people separately
If the estimates differ by a factor of three, nobody knows — and that is a more useful finding than either number.
-
Look at your last onboarding
How long did it take the most recent new engineer to become independently productive? A handover to a whole new team is at least that, usually more, and without the people who could answer questions.
-
Run the readiness check
The number of items in “not confirmed” and “needs checking” is a reasonable proxy. Each one is a thing the incoming team will have to establish for themselves.
It is nineteen questions and takes about ten minutes.
Questions people actually ask
They are exactly as accurate as the figures you enter, and the arithmetic is printed on the page so you can check it. It is an illustration of a scenario under your own assumptions, not an estimate of your situation.
No, and it never will. Nobody can know what a delay would have cost you or what would have happened otherwise. This page shows what the assumptions you entered imply, and stops there.
Because it is usually several engineers for several weeks, and because it is invisible in a budget — it appears as ordinary salary rather than as the cost of a transition. That is precisely why it gets underestimated when a handover is planned.
It varies enormously and anybody quoting an industry figure is guessing. The one thing worth doing is asking two people on your own side separately: if their estimates differ by a factor of three, nobody knows, and that is itself the finding.