Project Management for Founders – Effective Planning (Part 2)

In the previous article, I described how, as a Founder, one could relatively quickly and easily start building a useful plan. This took the form of a network of connected outcomes that illustrate what needs to be done to achieve the desired end state, in the example’s case a wedding.

It looked like this:

I finished by saying that there was a mistake and wondered if anyone might spot it? In fact, I then realised that there were some missing links too. This is a prime illustration of the power of this approach – it is easy for non-project people to understand and constructively critique.

The mistake was that it looks as if the guests at the wedding miss out on the reception and in fact get involved in the post-reception cleanup.

I also realised that one cannot book the reception or the honeymoon until you know the date of the wedding (i.e. book the church).

Making those revisions brings us to this where for now the changes are highlighted in red.

Of course, the reader will say that there is a lot more to do than this, and that is most certainly true, but that can be added once the overall plan looks sound. It will most likely be achieved by taking each item piece by piece and decomposing it into its own network of outcomes.

For instance “Guests invited” may need to include

·       Draft invite list

·       Select stationery

·       Order stationery

·       Write invitations

·       Post invitations

·       Collect responses

·       Chase non-responses

·       Etc.

But we are getting ahead of ourselves and the “plan” above is a start, but only a start.

Now we need to make sure the “plan” is as simple as it needs to be, ie it avoids unnecessary complexity. Complexity is usually indicated by a complicated network of linkages or by strange paths for the linkages. Paths should cross as rarely as possible and lead from left to right.

The reason for simpler plans is that they are easier to communicate and execute and in my experience, have a better chance of success. That does not mean dumb it down to the point it is useless, but rather consider is this the minimum meaningful plan (note the play with MVP/MMP?) needed to achieve success.

There are three techniques I employ to achieve this:

1)    Improve layout; consider if moving the activities (and their links) around to make the picture look simpler!

2)    Remove redundant links; if one finds a “nest” within the links consider if are really needed

3)    If you are still not happy, consider breaking an activity into smaller pieces (last resort)

As an example, I might have a first pass plan that looks something like this:

On inspection, the following links are redundant:

·       A to D, because the limiting path is “A to C to D”, ie D cannot start before C is finished

·       D to Finish, because the limiting paths are “D to F to Finish” and “D to G to Finish”

Additionally path B to E goes right to left and C to D is flat. 

If I remove the redundant links and rearrange the plan it looks something like this:

Which I am sure you will agree is simpler.

If I take my own advice and look back at the wedding plan I may well come up with this:

Simpler, eh?

Now we still have no idea of how long a task, let alone the whole project will take. There are a number of ways to do this. One of the most obvious is to calculate how much work needs to be done, consider how many resources you can assign and how much time they can give to the task; divide the latter into the former (i.e. work/effort), and you will have an estimated time. The trouble with this is that it takes quite a bit of effort to gather all the numbers and even then it does not in itself allow for the fact that it may be 2 man-days of effort, but take 4 weeks to complete, because of postal services etc.

A better approach is what I call the risk-adjusted, Delphic approach. Delphic, in this case, means taking advice from people with better expertise/knowledge of the undertaking who can give you a real-world estimate of how long each piece will take. Of course, you can consult more that one expert and come to some blended estimate if you prefer.

The “risk-adjusted” is an attempt to ensure there is a degree of reality in the estimate. As Founders, you are almost certainly optimists and when talking with others I know you want to give the best impression. This is over an above the usual human optimism.

The truth is estimating project tasks has some nuances. The first is that the most likely thing to happen is that a task takes longer than expected. Of course sometimes that is not the case, but usually it is, but as humans we tend to gloss over the worst view of things.

If a task (requiring reasonable effort) is estimated to take 7 days that is a figure you might use, but it is just that a single figure. It might that if everything goes right and the stars align you might be finished in 3 days, but then if things go wrong, supplies are delayed, you have to rework aspects, etc. it could take 21 days. So how do you reflect that?

Well, a good technique which approximates a statistical distribution is the following

 ((1 x Best Case) + ( 4 x Likely Case) + ( 1 x Worst Case) ) / 6

In this instance (1x3) + (4 x 7) + (1 x 21) equals 52. Divide 52 by 6 you get 8.66 days.

Now this is better than the single estimate of 7 as it reflects the natural uncertainty in planning, but even now I would suggest spurious accuracy and rounding (because your pessimism is probably still optimistic) to the nearest sensible unit – in this case 9 days

So back to the Delphic approach, ask the people who are likely to know what they think the best, likely and worst cases are. From those that the data you want, maybe an average, and apply the 1-4-1 approach to estimate a duration for each outcome.

Next time we will look at plugging those back into the plan and finding something called the critical path.