Table of contents
- Why estimating Agile projects can be more challenging
- The ten most common Agile estimation techniques
- How to get started
As a project manager or business owner, being able to accurately estimate work is one of the more challenging parts of the job.
There’s the financial aspect, the projections for future quarters, and the need to allocate resources so teams are properly utilized.
All of that comes down to having the right Agile estimation techniques.
When estimating is done well, projects are done on time, businesses are profitable, and team members are happier because their deadlines are attainable and their priorities are set.
Traditional project management methods look at estimating in terms of “bottom-up.” You spend time upfront making the schedule, planning out the tasks and deliverables, and then you break down each one and estimate hours and costs starting from the end.
From there, it’s up to you or the project manager to keep teams on track both in terms of deadlines that are already set, and hours allocated for each deliverable.
Agile project management, on the other hand, doesn’t quite work this way.
When you’re estimating Agile projects, you flip the order and use gross-level estimation. You’ll start with a broad estimate for different parts of the project, and then continually refine as more information is available.
Much like Agile planning, estimating happens continuously and gets more detailed as the project progresses.
Because it requires a different way of thinking, we’ve compiled the ten most common Agile estimation techniques so you can compare and see which one works best for your team.
You’ll also want to gather some tools before you get started, including:
Pen and paper
Sticky notes
Clear wall space or table
Your list of items, user stories or tasks
If you’re using a work breakdown structure, you have that last bullet ready to go.
When it comes to deciding which technique to use, consider:
The number of items or tasks you have to estimate
The size of your team
Co-located or all in one office
Tools available to facilitate the process
Work style and personalities
You might also want to brush up on the Fibonacci sequence, as some techniques use these numbers or a modified version to help assign values to each task.
Now, let’s take a look at each technique for estimating work in agile projects.
The problem with committing to an estimate before any work begins is that they can be wildly inaccurate. Even if it’s the same team and the same type of work, estimates can be unrealistic. Especially if you fall into the trap of thinking, “We’ve done this before so we’ll be much faster the second time.”
That’s what makes three-point estimation so useful. You have three values: the most likely estimate, plus the optimistic, and pessimistic ones.
From there, you get the most accurate number by adding up the values and dividing by three.
A great example of a three-point estimate is with project deadlines. Let’s say you have a new client in need of a website. Maybe you’ve done a website design before for another client and it took a month. You might set the realistic deadline at four weeks, the pessimistic deadline as eight weeks, and the optimistic deadline at three weeks.
First, total up the work periods:
4 + 8 + 3 = 15 weeks. Then, divide this by the number of points (three) in your estimate:15/3 = five weeks. A good, realistic estimate for your project is five weeks.
Best for a small number of items, planning poker is a useful technique that gets unanimous buy-in quickly. Each team member gets a set of specially numbered cards, discusses the requirements for the item, and then submits their best estimate anonymously. Discussion happens until consensus is reached.
This approach is ideal for 10 or fewer tasks, and teams of up to eight people.
We’ve done a lot of planning poker on the Hubstaff content team. On our Sprint calls, we sit down to assess how many story points (between one and five) we should assign items like blog posts, editing, adding posts to our CMS system, emails, and other content we create.
Planning poker has worked well for us because it’s hard to make sure each member of the team has an equal workload with such a wide array of tasks. We’ll never be able to capture how many marketing emails could be written in the time it takes to write a blog post, but planning poker helps us come pretty close (and it’s a lot of fun, too!)
This estimation approach works by having team members group similar items. If tasks seem related in scope and effort, you put them together until you have a clear set of groups.
You can use the same set of values as other techniques (Fibonacci sequence), or make the groups more broad so it’s closer to the large, small, and uncertain method.
You’ll see a lot of affinity grouping in development because it just makes sense to group similar work. You might sort work into front-end development, UI/UX improvements, site/app maintenance, and more. Or, as mentioned above, you might want to group all feature updates by scope and effort as opposed to which aspect of the app they’re related to. The possibilities are endless.
Also known as an ordering protocol, this technique places items in an order from low to high. Each person takes a turn, choosing either to move an item up or down one spot, discussing an item, or passing.
Your process is done once everyone passes on their turn.
You might see Random Distribution happening a lot when trying to decide which new features to work on. By giving everyone a turn to shift priorities around until each person is satisfied, you create universal agreement on which features to prioritize (and therefore, which features to give the most bandwidth).
XS, S, M, L, XL are the units you’ll use to estimate Agile projects for this technique. One of the reasons this approach is successful is because it’s a departure from standard units of time, and thus, can help teams think more critically.
Your estimation unit could be anything here, and thinking outside the box can help your team objectively compare items for better estimates.
A good time to use the T-shirt sizes concept is with video creation. A quick Instagram reel or Tik Tok might be an extra small, while a live product demo or webinar might be an XL. There’s plenty of video content that would fall in between, too.
Like planning poker, the bucket technique aims for consensus through discussion and assigning values to each task. The facilitator starts with one task, sets it in the middle, and then continues to read and place tasks in buckets relative to the first one. The team can divide up tasks and bucket them before coming back together and reviewing.
Discussion is key to make sure everyone agrees before the final estimates are set. You’ll also look to make sure tasks are evenly distributed and not, say, all under the 200 bucket.
Like planning poker, t-shirt sizes, and other point systems, assigning points to a “bucket” of tasks applies to virtually every profession. One example might be in the graphic design world. Designing a simple social media graphic might fall into a 100-point bucket where designing a brochure might fall into a 300-point bucket.
Fast, simple, and productive, this technique is like bucketing but with only three possible values to assign. You’ll start out discussing and then divide and conquer to get all the tasks added to the large, small, or uncertain groups
Just like with buckets and other methods on this list, the large, small, uncertain approach is very versatile. One example that comes to mind is in the CPG industry. If you work in a seasonal role, you know that projects around your Q4 holiday catalog could fall in your large bucket, the summer holidays can sort into your small bucket. Newer products not tied to this seasonal production line could be considered uncertain.
This one is fairly simple: each person gets a number of dots and uses them to vote on which projects are big and small. More dots mean more time and effort is required. Fewer dots indicate a fairly straightforward and quick item.
Dot voting is also great in the development world. Similar to random distribution, a great example of dot voting would be in planning new features. Maybe you assign more dots to features that existing customers have requested. Maybe you assign less dots in areas that would be more expensive and likely wouldn’t yield mass approval from your customer base.
Initiating a top-down estimate involves facilitators communicating the project's fixed deadline to the participants. Participants form groups to segment the project timeline into distinct phases. After allocating time to each task, team members meet to cross-check their estimates.
A top-down estimate could be great in the e-commerce space. For an e-commerce apparel store, the phases are already pretty clearly defined. You can break projects into phases like product development, design, and distribution. Since all clothing you sell will fit nicely into this model, iterating on the top-down estimate method can help you become increasingly accurate in how long it takes to sell new products from inception to point of sale.
When adopting a bottom-up estimation approach, the process begins with meticulously evaluating the time required for each project task. These individual task assessments are then consolidated to determine the time necessary to complete all project phases.
This methodology not only aids in pinpointing tasks with extended durations and empowers team members to devise strategies for improving overall productivity.
Unlike with top-down estimations, bottom-up estimates happen before you break ground on a project. A great example of this might occur in high-pressure work environments like commercial real estate. While e-commerce stores often begin as Start-ups and benefit from scaling along a top-down estimate style, enterprise real estate companies might use bottom up to predict the cost of a new condo complex before they break ground on it.
Which one will you try first? If you prefer, you can use a different method at the start of any roadmap or quarter.
Using an Agile project management tool can give you a clear list of tasks for estimating.
Hubstaff with the Tasks add-on helps you plan Epics with individual tasks that can be assigned to team members.
Add due dates, comment on tasks, and know where every item is at with Kanban-style project boards. From there, you can plan your Sprints to prioritize the right work for your team.
It’s a simple, visual way to manage Agile projects.
Hubstaff Tasks keeps work on track with Stand-ups, Sprints, and more