How to Prioritize Projects and their Tasks
As a leader of a marketing organization, I am in a constant state of juggling competing priorities. My clients each believe their priorities are the most important, and thus should be completed first. In reality, that is almost never the case. In fact, many times their priorities are the least important. It's my job as an executive to understand the goals of the organization and than translate those into priorities. I must rank the projects in order of how they will help the organization achieve their goals. At the same time, I have to balance those priorities against team retention and my career. These often do not correlate evenly. In other words, if I have a stakeholder in a department request a project that I know adds little, if any, value to the organization, I also have to understand the pros and cons of completing the project. If I complete the project as requested (sooner than it really needs to be completed), I might win favor with that leader and department. Doing so could advance my standing and that of my team's. Conversely, delaying the project beyond the leader's timeline (though it might be justified), could hinder my advancement and that of my team's.
Take a team of 30 in an organization of about 2,000. There can many concurrent projects - well over 100 marketing projects. Each project could have one task or hundreds, but usually somewhere in between. In a large recent project, whereby I wanted to overhaul a web property, my team did our standard research, due diligence, collaboration with stakeholders, and cursory brainstorming, to yield about 25 enhancements. Thus we had our project, "Web Refresh" and about 25 sub-projects or tasks. We needed to figure out how to order the 25 tasks so that we could deliver the most value first. We leveraged the 9-Square Prioritization Matrix. If I wanted to be cute, we could call this the Soschin Priority Matrix. Here's how it works:
- Each task is assigned a rating for level of effort of HIGH, MEDIUM or LOW. I use two scales (though you can only use one scale at a time, so chose the appropriate one for your project.) The first scale is for large projects. HIGH level of effort (LOE=H) is for tasks that take more than a day but less than a week; LOE=M is for tasks that take about a day; and, LOE=L is for tasks that take less than a day.For tasks that take longer than a week, adopting a concept from the agile method, you need to break the task down into more sub-tasks. There's an optional scale for very small projects that I rarely use, but it comes in handy sometimes and that is H=more than an hour; M=about an hour; L=less than an hour. It's very important to craft your ability to estimate task LOE so that over time you can do this accurately, accounting for research, experimentation, iteration, testing, refinement, etc. Underestimating does no one any good.
- Next, assign each task a value for return on investment (ROI): HIGH, MEDIUM, or LOW. HIGH tasks should be easily measured and translate into significant revenue opportunities or expense reductions; LOW should be assigned to tasks that round the business out, fill in gaps and are more than wastes of time; and, MEDIUM tasks should be somewhere in between.While my explanation of ROI can seem a bit arbitrary, that's simply because I do not know your business or your project. In other words, you should be able to figure this out on your own. If you cannot figure out if a task has ROI, perhaps you should not be working on it - stick in the the "blue sky" or "rainy day" task list and let it marinate (or fester and die.)
- Using the table below, score each task according to the assigned LOE and ROI in steps 1 and 2, respectively.
At this point, each task should now have a value of 2, 4, 6, 8, or 10. You're now much closer to a ranked list of tasks, but you've also probably noticed that some are colliding with one another, in that a medium LOE with a high ROI (score=8) has the same value as low LOE and medium ROI (also score=8). That's okay... keep reading.
As you start to rank your tasks by score (highest score is the first task you should complete), when you have a tie, your knowledge of the project and the specifics of the task will be the tie-breaker. If you cannot figure out which one to do first, flip a coin because it doesn't matter.
For those of you who belong to Mensa, and for extra credit, you can add a third-dimension called "PCW" which stands for political capital won. It's a similar scale of 1-3-5, with low PCW being 1 and high being 5.For example, if you have an 8 (LOE=M + ROI=H) and the PCW is high, the task would get 5 more points for a total of 13. Note, that by adding the 3rd dimension, you go from 5 possible scores to rank tasks (2, 4, 6, 8 & 10) to 12 possible scores (the original five, plus 3, 5, 7, 9, 11, 13 and 15). So, on projects with lots of tasks, if you really want to get specific, go with the 3rd dimension. But that can really over complicate things. In some of my former roles, there was a lot of PCW at stake, so the 3rd dimension was in play frequently.
So that's it... that's how you translate a complicated set of tasks into an organized and methodical plan of attack. It's a great way to communicate to your project team as well. Over time, by completing this exercise, you team should get more accurate at agreeing on ROI and LOE. I have found that when I introduce this in projects to new team members, folks tend to group tasks into more narrow scores, such as lots of medium LOEs and ROIs and therefore a lot of 6's. That's okay, but over time you should be able to spread the tasks out over the matrix. There's no formula or bell curve for doing so - it just depends on the project.