An Effortful Estimate
03 May 2026
Coming Up With an Estimate
Estimating how much time I needed to complete a task was often very tricky, as sometimes I wouldn’t fully know how to complete the task, so I would spend some time figuring that out which made the task take longer. In addition, there were several times I ran into other issues that I had to fix before completing the task or I had messed something else up when completing the task. That being said, I did have some estimates that were pretty far off the actual time it took, but most of my estimates were relatively accurate.
What I Learned from an Estimate
In hindsight, I think the estimates I made in advance definitely helped give me a better understanding of what I should think about before making my estimate, as at the beginning I was kind of just looking at the task and slapping an estimate on without thinking. But over the course of the project, I learned to account for more factors in my estimates and learned about how long certain issues can take. For example, there were several times in our project where we modified our data models, and we would have to make prisma migrations. Sometimes we would forget to apply these migrations or accidentally overwrite them, which caused some delay in our tasks. So, after knowing this, I slightly elongated some of my effort estimations that involved modifying data models, as a little more time would be allocated to making sure the migrations are in place.
Tracking the Task
To track my tasks, I just ran a stopwatch and recorded the minutes/hours it took for me to complete the task. I would say my tracking wasn’t super accurate, as there were probably some times I got distracted and didn’t stop the timer, or I rounded the time off at the end. There were also times where the task took so long I had to take a break to preserve my sanity, so I didn’t get super accurate times there. I also didn’t really track the non-coding time and coding time too accurately, as I would often switch between the two, which made it tricky to track.
What I Would do Differently
Next time, I would definitely try to track my coding and non-coding time more accurately, as well as making better estimates depending on the task. I think underestimating a task is probably better than overestimating, as if you do complete it quicker than your estimate, you have extra time to use. Overall, I think my estimates were accurate enough to be useful, but definitely a learning experience for some tasks. By using past effort estimations, I can make more accurate estimates for tasks moving forward, which will help manage my time and prioritize tasks that may be more difficult to complete.
AI was not used to write this essay.