Error correction time allocation

Question

We asked the client to give us an estimate of the time for each error that we have.

Although we have an established schedule for fixing errors and time is allocated for it, we do not have a time distribution for each of the errors that we have. We simply set the priorities for our mistakes and ensured that the highest priority errors were fixed in the allotted time.

I'm not a fan of taking the time to make mistakes, simply because:

• This is usually inaccurate. It is very difficult to understand how long it will take to fix it.
• Time waste.
• Affects code quality
• Creates more mistakes in the long run (we may miss some things in our attempt to complete it by the deadline).

How should we solve this problem when we do not want to indicate the number of hours for an error, but only the time frame of which errors will be fixed?

How do you set aside time for your mistakes? Is it effective? Is it worth the time and effort?

Answer 1

The only answer I can give is to be extremely conservative. Guess how long it will take, and a few of your four assumptions. Use this as your rating. As you said, it’s very difficult to understand how long it will take to fix it, and it’s better to say that it will take more time than actually than to be caught “violating your deadline” because you are not conservative enough.

Answer 2

The company I often work with often receives unreasonable requests from our customers. The main thing to remember is that customers want to be in the know. We found the best way to do this in terms of status reports.

So, first we explain our position very well. In your example, it would be something like this:

We have a set of schedules for fixing errors in our project, which we historically know well about staying on schedule. However, the process of detailing how long each error will be fixed is quite error prone. We will be happy to provide you with weekly updates (or twice a week or daily depending on the client) about the corrected errors and verified corrections.

However, I find it helpful to try to evaluate how long each error will be fixed. The reason for this is because you need to understand what is the total time to fix all errors. You will not be able to get an accurate estimate if you do not have an estimate of how long the individual parts will be fixed. Of course, these can be rough estimates (calculated no more than the cost of researching the problem) - you do not want to spend too much time evaluating. Then I usually include an extra 20%. So say grades for errors of 3 days, 5 days and 2 days. Then I inform the client that we can fix the errors in 12 days. Then, of course, you may need to add more time to test and repack your product before you can enable them to deliver.

Answer 3

Do not think about it in terms of evaluating how long the errors are fixed, because you cannot correctly evaluate it.

Think about it in terms of managing customer fury . If you tell them that the errors will not find time to fix, and they will ultimately take 3 months, your client will be happy with you now and furious with you in the future.

If you tell them that it will take 3 months to fix the errors, and they really will take 3 months to fix (what they will do), your client will be furious now and happy with you in the future.

I usually say that mistakes do not take any time at all (2-3 days, it seems, is a good pacifying number).

Answer 4

It should coincide with the assessment of any other task that you have. Divide it into the smallest possible tasks and evaluate them as accurately as you can with a supplement for the unexpected. Then give them a range so that you are not attached to a specific date in tasks that were not clearly defined. There is no difference between estimating time in order to correct a mistake and estimating time for implementing a function with vague requirements.

Answer 5

You are right, estimates are usually inaccurate.

Perhaps you want to ask them how much each error will cost them if it is not fixed. You can then do the corresponding calculations to find out if they should ever be fixed, and how much time you (or really, they) can afford to devote to each error.

Answer 6

Why not just select multiple ranges for the severity of the errors, for example. 1 hour, 1/2 day, 1 day, 1 week and appoint against them. As a rule, you will have a feeling that the error is the one you don’t know about, attach the worst figure to it!

I would not have thought that you would need to be assessed at some higher level than for the reasons you quoted (too much time for investigation, etc.).

I do not think this is a waste of time. Your client wants to know more than the number of errors and their priority - they want to feel how much work remains.

Under no circumstances will this lead to more errors being created. You do not have to rush the clock to fix it. If you rated 1 day and it took 10 hours, that's fine. If you rated 1 week and it took 2 hours, a good result!

This is just an assessment exercise!

Answer 7

Usually we agree on which errors should be fixed for a particular version, and then determine the time frame for fixing all errors. For each individual error, there is a lot of uncertainty / variability in how long it may take to fix, but it tends to average with a lot of errors. For some errors, which, as you know, will take longer, it may be possible to give some estimates, for example. if you need to write a simulator or test environment for it.

Answer 8

If these are errors that have been found and reported, you should be able to develop an estimate of the time to fix it (and the time for retesting). Confidence in the valuation is likely to be proportional to the time you spend evaluating, perhaps explain this value to the client.

If there are several related small error reports, perhaps you can collapse them into a single omnibus report. This can lead to the fact that the client will try to choose which errors to correct based solely on individual assessments.