Mainly based on previous experiences.
I know from previous projects how much work it is about to implement a particular feature, and that estimation is more accurate as I am more aware of the project.
When estimating the time required to implement a feature, you will first learn how to build that feature.What steps do you have to put to build it?
Then you’ll check if you can relate it to something you’ve built before, and check how much time you lost.Even if you do not have that experience, you can often still give a reasonably accurate estimate.
If you want to calculate the length of time (approximately) for a larger project, it helps when you do the estimation with a whole team.However, this can only be made when the entire project is designed and divided into userstories that you can start estimating. The duration of the project is then the sum of the development time of the userstories with extra time for overhead, testing and bugfixes.
Simp芒 鈧?娄 that suck you out of your thumb!
But well, experience with previous projects should give you a reasonable indication.In general, you do this by first dividing the project into as many small parts as possible. Then you can try to estimate how long each part needs about. Count together and add 50% for runoff and unexpected cases, and keep in mind those parts that you can’t accurately estimate how long they need.
However, making a good estimate remains difficult.But by splitting the project into small parts you can determine which parts have priority. So you develop first. If you are short of time then you can omit parts to be ready on time with the project.
There are several tools that allow you to make such schedules, although it can be in Excel as well.Important here is that you keep an eye on the progress to check that you stay on track. If not, you sometimes have to take tricky decisions.
Something I have had to deal with in the past was that I had to create an application, but the customer would take care of images.This was a week’s work but at some point I needed those images that the customer still hadn’t delivered. So then a dummy plate is used and the customer thus delivered a product containing dummy plates and the remark that they should have supplied material on time. And so the customer finally had to pay for the extra work when they finally came to the footage three weeks later. I kept myself within the agreed deadline.
But there may be plenty of other factors to eventually mess up your schedule and you can’t take that into account.Power outages, sickness or other delays can not be added, hence the extra 50%. Past experiences should make it easier to make estimates.