Can you out-puzzle a difficult project?

Sometimes when your project is in such a mess, it can be hard to figure out what the next best thing to do is.

Have you seen those puzzles that are small irregular shaped metal tubes all intertwined. The premise is that by putting them into the right positions, through the right sequence of carefully planned movements, all the pieces will separate. But until you know those movements, the pieces seem impossible to separate. You start using brute force but it doesn't seem to make a difference.

A project in a mess feels a lot like this. You know what good looks like (all the pieces separated), you know what you've currently got (a mess), but you can't work out the steps to get from current state to target state. And just when you think you're getting one piece untangled, your next move seems to make the whole in an even bigger mess than it was before.

Let's untangle

The process to resolve this is actually not that hard. As often in projects, we start at the end. What does good look like?

I like to be as prescriptive as I can. Take each discipline and describe in detail the standards expected. Finances - I need to be able to track spend to date and forecast spend. Requirements - I need them documented, signed off and in change control. Risks - I need a mitigation plan for each one with an action owner and a next action date. These are not difficult to define.

Take inventory

Work out where you are today. Finances - not trackable. Requirements - not defined. Risks - not universally mitigated and actions not progressing.

Next, I give each area two ratings.

The first rating is where we compare to where we want to be. The score for each area is the number of steps needed to fix the problem. Risks need mitigations everywhere (count 1), action owners everywhere (count 2), a next action date (count 3) and a weekly meeting to review progress and blockers (count 4). So Risks score the 4th number in the fibonacci series (1-2-3-5). Risks scores 5.

The second score I give is to rate the size of the negative impact that could be occurring because of the deficiency. Again, I like to score out of 10. But you can RAG, or t-shirt or any other system. The idea is to identify the focus areas. Lets say risks score 6 out of 10 for impact.

Make the first move

Next, I create a combined score. Some way of showing that something that is 1 step away from target is probably less of an issue than something 8 steps away from target. But if the 1 step away is a big impact, maybe it's a bigger deal.

The way I do this is first to multiply the 2 scores. 5x6=30. Then I work out what the score would be if I were 1 level lower on the fibonacci series. That would make this 3x6=18. The value of taking the next step is the difference between those two scores. The value for taking the next step on risks is 30-18=12.

Repeat for all your problem areas. Define the steps, create your scores, score the impact, calculate the score for taking the next step. Once you're done, review all your next step scores and do the one with the largest score.

If you follow this disciplined approach, it won't be long before the mess is feeling like less of a mess.