[edoceo]

I don't estimate, I work the other direction.

How much is it worth to fix? That's a much easier value to discover.

From there work backwards to determine how much to spend and where to prioritize.

Eg: a bug generates 3 email and 3 phone support issues per day, at a cost of 1h or $33/day - $12k/yr. Easy to justify spending one dev-week on the issue ($6k).

Or missing feature that lost a sale of $50k easily justify $5k to make some progress.

[lukasLansky]

Funny, I find measuring opportunity costs much much harder than measuring engineering effort.

Even in your example, are those phone calls the substantial costs, or is it customer confidence in your product being lost the main motivation you should do something with the bug? If so, what's the dollar value of this confidence? Also, do you have monitoring good enough in place to see how many customers are hitting the problem every day without telling you? Do you think these customers are less or more frustrated than the one they are calling you?

Can you attribute lost customer to a particular feature not present, to a particular bug the customer experienced just before they refused to deal with your software anymore?

[edoceo]

A lot to respond to, I'll try.

I'm measuring stuff that is actually happening, whatever it is (phone, email, error log, metrics, etc) and have to assign a cost. I find that easier than guessing how long to discover the right answer to a (maybe) complex problem.

For phone and email support it's easy, directly measurable cost. One would decide for them selves how to factor those intangibles (ie: confidence).

Currently I do things like this-ish: - Phone and email at direct cost; but if the issue is a duplicate double cost. - Lost deal factor at 15% first year value - Logged errors at $10/first, $2/dupe - make values that match your company revenue/expense.

We try to exit interview clients but generally have a good feel for it - ours is a pretty high touch relationship

[jwdunne]

I think in the first example it's easier to calculate based on the time spent vs the salary + overheads of the employee(s) dealing with the issue.

If something took up one hour of my day and I could spend a week one-off to reduce that to 0 hours a day, that's a no brainer.

[edoceo]

As always: https://xkcd.com/1205/