How to measure time to ensure your projects aren’t late

There is a fallacy in every project management tool I’ve seen. And the fallacy bleeds into how we run our agencies, consultancies, and freelancing businesses.

It is the assumption that a day worked equals a day of progression on a project. i.e. 1 calendar day = 1 project day.

We all intuitively know that the time you estimate does not correspond to how long a task actually takes. For example: your developer might estimate that a task will take 3 days, but you know it will probably take two calendar weeks before the client can see it.

Let’s explore the gap between estimated time and calendar time, what we call effort time here at Clockk.

What is estimated time?

Estimated time is the amount of time you think it will take to complete a task.

What is calendar time?

Calendar time is the amount of time that passes from start to finish, including any distractions, meetings, breaks, weekends, vacations and sick days.

What is effort time?

Effort time (in Clockk’s view) is the time you spend directly working on a task or project. This includes actual implementation (your work in Photoshop or VS Code or Google Docs) as well as meetings directly related to the project.

In addition to estimated, calendar and effort, we often come across actual time, elapsed time, and duration in project management. These three terms are ambiguous. “Duration” is any amount of time, whether estimated, calendar or effort. “Elapsed” refers to the time that has passed since the beginning (of the project/task/etc.). “Actual” time can be either the calendar time or the effort time, depending on your perspective. We try to stay away from these three terms to reduce confusion.

Why is my project late? A micro-example

While writing this article, we actually had a great opportunity to micro-test our thesis. I asked one of our developers to set up a test version of the Clockk application. This is how it turned out:

• Estimated time: 1 hour (his estimate) / 2 hours (my estimate)
• Effort time: 2.5 hours
• Calendar time: 7 business hours (the functionality was live the following business day)

The gap between estimated and effort time was pretty small. The big gap is calendar time. What were the causes?

• The developer in question is our most senior developer, so he is often interrupted with other tasks (in fact, this request was an interruption of another task, which was itself an interruption in another!)
• He is also assisting a junior developer come up to speed with our infrastructure.
• Some of the task included waiting for servers to start and caches to expire, during which time he returned to his previous work task.

On a small task, we see a 1-2𝗑 difference on estimate to effort, but a whopping 4-8𝗑 difference between estimate date to calendar date. If the ratio holds, a 2-day (estimate) task can take 2 to 3 weeks until the calendar date where the work is delivered.

On one hand, it’s staggering — how could a two-day task take 3 calendar weeks? On the other hand, it’s very familiar, isn’t it? 🙄

Why are my estimates wrong?

“Estimates are always wrong” is not as true as we believe. In fact, effort estimates are probably more accurate than you think. It’s our ability to estimate calendar dates that’s in question.

• We fail to adequately account for the distractions during our work days. If a worker is sufficiently experienced (i.e. needs minimal on-the-job time for learning), 100% dedicated to a single task on a single project, and free from meetings and interruptions by peers and management, their estimate, effort, and calendar times should align fairly closely. This is why remote workers appear to be more efficient than their in-office counterparts.
• We do not track or treat a worker’s utilization rate (the percentage of a worker’s time that is billable) as their capacity to work on a project. Most agencies and consultancies fail to track individual contributors’ utilization rates. Those that do often use utilization rates as targets to achieve and not as reflections of that individual’s capacity. Instead, we should determine each individual’s utilization rate in a judgment-free way, and use that value as a multiplier to determine the individual’s capacity to do work on the projects they are assigned to.
• We don’t measure the total concurrent task-and-project load on an individual. If the team member has a single task assigned to them, they can devote 100% of their capacity (post-utilization) to that task. If they have five concurrent tasks, then all things being equal, each task gets 20% of their capacity. Since humans can only work on one thing at a time, they must divide their time between tasks, addressing the most urgent or important tasks first. The other tasks necessarily wait, increasing those tasks’ calendar time.

The important thing to note is that in all three scenarios above, only the calendar time is impacted. The effort remained the same, whether one task or many, regardless of utilization.

How can I get my project back on track?

The simple answer: If you want a task delivered in the same calendar period as estimated, then don’t distract your team with any other requests. Don’t even ask them how their weekend was. Let them devote 100% of their effort to the task.

The realistic answer: You can’t change reality. Our [work] lives are filled with competing priorities and we all have to do our best to balance those priorities and get things done. Instead, quantify each individual’s load, to determine their capacity to work on a particular task.

1. Determine each individual’s utilization rate. As best you can, be judgment-free. A low utilization rate could mean that the individual is often helping their peers and being asked about projects past and present. A high utilization rate could mean that the individual is working inefficiently or too many hours. Use the utilization rate to determine the individual’s maximum capacity for project work. i.e. if your designer has a 73% utilization rate and they work a 37.5-hour week, their maximum daily capacity is 5.5 hours. A 6-hour task will take at least 2 calendar days to deliver.
2. Determine each individual’s load. The load is the count of concurrent tasks at any point in time. If they have 3 tasks on project A, and 1 on projects B and C, then their total load is 5.
3. Determine the individual’s capacity. The capacity is the amount of time they can spend per period (day, week, etc.) on a single task. Assuming the 73% utilization and 5 tasks above, your team member can realistically spend a little over 1 hour per task per day, or 5 hours per task per week (7.5 hours in a day 𝗑 73% utilization 𝗑 1/5 tasks = 1.1 hours). A 6-hour task can take 7 calendar days to deliver if that project gets the same attention time every day.

Obviously all tasks are not equal in weight or importance. In practice, if that 6-hour task is important enough, it will be delivered on the same calendar day, but the other 4 tasks will wait until tomorrow, even if each is only 1 hour long.

Overcoming the differences between estimate, effort, and calendar time

Every day, freelancers, project managers, business owners, and more are making prioritization decisions. We choose one task over another, and one project over another. We don’t trust our estimates (although perhaps we should!) because they never align with our calendar delivery date. This is because we don’t quantify or reason about the effort that tasks take, especially when we work on multiple projects in the same day.

Most project management tools treat projects atomically — they exist in complete isolation from one another. Projects may exist alongside other projects, and they may have the same team members assigned to them, but there is no way to manage that individual’s time distribution across projects for any given period. Furthermore, project management tools are idealized and forward-looking whereas time tracking tools are backward-looking. This means that we have no way of knowing, ex-post, whether our prioritization decisions were good ones.

We need a tool that will allow us to see how we are spending our time and make better decisions about what tasks to work on.

Clockk's AI-powered tracking with no start/stop timers helps you nail down the effort time of a project — including all those little bits of work scattered through the long tail of winding down a project. Use Clockk’s ridiculously accurate data to see how much time you actually have available to do work.