Sloan's Inbox: Any advice for estimating work?

A piece of advice: This is hard to get right, don't assume anybody else is doing it that well.

But also don't treat this is useless busy work. It's a valuable exercise.

As far as advice for doing it better: It helps to be 100% on the same page about what factors other people in the exercise are considering, this is ultimately more about team members being on the same page and working together vs just being accurate above all else.

Try to estimate everything. By 'everything', I mean tasks ranging from how long it will take to brush your teeth to how long it will take to how long you will work on book. Don't estimate complex things all at once. If you want to estimate how long it will take to finish your work, divide it into smaller tasks and estimate them individually. The more you estimate, the more accurate you become.

• Pick a random number
• Multiply by a magic constant

1. Don't just give a single date, give at least two: a 50% confidence estimate, and an 80% confidence estimate.
2. Via experience, learn roughly how wrong your estimates are, and take that into account for future estimates. For example, if you find that you underestimate everything by around 50%, then double all of your future estimates.
3. When in doubt, be conservative. It's a lot more common to underestimate than to overestimate. There's scope creep, people get sick, there are bugs that are hard to track down, etc etc. There's ALWAYS going to be more work than what you can foresee.

1. Create an Excel spreadsheet.
2. Write down every feature needed in the app in the spreadsheet.
3. Try to estimate how many hours it will take to complete each one. You don't have to be perfect at it, just try your best.
4. Sum all the hours up.
5. Multiply total hours by your hourly rate.
6. Multiply the number you got from step 5 by 1.5 for extra safety.

Always mention that the estimate is non binding.

Here is one of our blogposts that may provide more insights. Please see the "Know the Cost" section of the link below. We haven't published it in the Dev Community yet, so here is the original link:
Top 4 Reasons to Hire a Software Development Consultant

Multiply your original estimation by Pi, that way you always over deliver, and everybody is happy ^_^