There are loads of different ways to estimate, from looking back over past projects and using that information to help you work out what this new project might need in terms of resourcing and costs to detailed modelling techniques and more.
You probably use a bunch of different ways to estimate, depending on what you are estimating and how much information you have about the thing being estimated. But did you realise that estimating techniques fall into three categories?
Two of those are the ones you probably studied on your project management courses: quantitative and qualitative. The third is a category that you might not have come across unless you work with agile or iterative techniques: relative estimating. Let’s have a look at each of those.
Quantitative (I hate typing that word!) is a way of estimating that relies on you having data and numbers to be able to make a definitive, mathematical estimation of how long things will take or how much they will cost. As the name suggests, this type of calculation is based on quantities. If you know how many of something you need (resource hours, bricks, units, etc) then you can work out the cost because you know (or can find out) the price for a single item.
Of course it’s not always quite that simple. If the cost of a brick is 1p, and you need 100,000, you might get a discount on such a large quantity. But basically, these types of estimating are quantity- driven.
Types of quantitative estimating include:
- Analogous estimating
- Parametric estimating
- Bottom-up estimating
And you may have other techniques you use on your projects that rely on the same kind of approach: find the numbers that relate to the tasks and use them to extrapolate the total estimate for the project based on amalgamating all the figures for all the tasks.
Qualitative estimates are data-led. You don’t necessarily have quantities, so you need other ways of working out the time/cost. For example, how long will it take for the subject matter expert to update the policy? There’s an element of research in there, plus an approval process, and the time to write up and publish the new policy. That’s hard to quantify in terms of hours because it’s knowledge work, and we don’t use our brains in simple to understand units. You might get a burst of policy inspiration in the shower, or you might need a couple of days to let the ideas mull around in your head before you sit down and write the updates in 20 minutes.
Types of qualitative estimating include:
- Expert judgment (that old favourite: the educated guess)
- Observation (basing the estimate on what you see)
- Interviews (another form of expert judgement: getting info from other people)
- Surveys (sourcing info from the wisdom of crowds).
These are all suitable types of estimating that are OK in specific circumstances, and we do need a range of ways to size project work.
Finally, we come to relative ways of estimating. They are relative in relation to each other – the project tasks. I wouldn’t say these were particularly reliable techniques for estimating project costs (beyond resource time if that is chargeable) because sponsors tend to like actual figures for budget forecasts, not a statement about whether an activity is more or less expensive than something else. However, that could be useful information for a prioritisation exercise as it would help them understand what they could get for their remaining budget.
Relative estimating, when used for sizing project tasks, is about comparing the effort involved in doing the work to other tasks also on the table.
Types of relative estimating include:
- Planning poker or T-shirt sizing
- Affinity grouping
I use T-shirt sizing with my own work, although I don’t truly work in an agile environment. It’s a case of looking at the tasks (which, in an agile environment would be user stories) and deciding whether the effort involved is big, medium or small. In a team environment, we’d be doing that together but as a way to structure my To Do list, I do the exercise alone.
All of these types of estimating have a place in your toolbox and they offer a selection of ways to think about and plan our work. It’s easy to fall into the trap of thinking you have to use the same estimating method for every task on the project but you really don’t. Estimating approaches should be tailored to the task/activity/item. For some, parametric will be the standout winner and the most relevant way of working out the time required. For others, you’ll be relying on expert judgement or using planning poker to get a view of effort. Pick and mix your approaches from the three categories and you’ll end up with a rounded, appropriate way of planning your work.
Which of these categories do you use the most? Let us know in the comments below!