Since September 2015, I manage the project portfolio of all custom development at a large petrochemical site.
When I coached other portfolio managers before, my vision was always to improve the end to end value stream, from the ideation until the release to customers.
I spent my first months getting a good grip on the current process. Off course it is important to keep business running before you try to improve things.
Their existing process was pretty well structured and contained these steps:
– Intake meeting
– Initial project document
– Project application
– Final project document
– Final planning
– User acceptance testing
– Turn into production
I believe a portfolio manager can have the biggest impact in the 5 first steps. You can help the business owner to shape the idea or propose alternatives based on existing software or re-use.
Often a new request comes from a departemental view and doesn’t take the rest of the company into account.
Only when there are no solutions available, do we start with a project application.
This is where the business owner works out his business case, high level scope and budget, risks, people.
I prefer to use a more lightweight approach, where the business owner collects just enough information for senior management to decide whether or not the idea is worth the investment.
That’s why it’s important to include the link to the company strategy in the project application.
Many companies use a framework based on categories that define if a project is worth prioritizing. For instance, legal obligations, business critical, …
Even then, the most important piece is the link to the company strategy. It will give senior management a better base to make a decision.
Every once in a while, a project application gets approved with the most urgent priority. “This project must start asap and gets priority over all the others in the pipeline!”
There are valid reasons why senior management pushes a project like this. Often business continuity is endangered or legal implications are imminent.
Still, I want to avoid this behavior because it has big consequences on the entire project portfolio.
Other projects get delayed, planning gets shuffled, commitments get broken. I want to reach a point at which I can guarantee the planning of the current quarter.
For projects that are lower in priority and automatically fall out of the current quarter, I don’t communicate a specific date, I only show in which of the next quarters they will be delivered.
It’s basically aknowledging that the distant future is too much subject to change and the portfolio will be subject to variability.
On the other hand, the current quarter has to be clear enough to communicate to the stakeholders. And for most stakeholders this is enough.
A key to keeping the portfolio stable is to make sure the projects are not too large.
If your project teams are occupied for months on a project, there is too little throuhput in the portfolio to avoid new high priority project being pushed into the planning.
On the other hand, if every few weeks, a new project gets started, management will be less inclined to push a new project but rather let it follow the normal flow.
A good technique to stimulate this is using smart release planning. Often, a new business idea can be split into several waves. For instance, the first wave could be a subscription landing page, the second wave a website with the functionality to support orders, a third wave extra functionality to support self-care, etc…
If no urgent new projects arrive, these waves can follow one another. If not, we can insert the high priority project in between 2 waves.
As a result, we already see an increase in delivery. More projects are being delivered, faster to our customers. At the same time, the portfolio backlog has shrinked from a 2 year- to a 1 year horizon.
We were able to reduce our headcount with 25% and still maintain the same service level.
The road is still long to create a true lean portfolio, but the first steps have been taken.