Sustainable technology with balanced portfolios

When delivering and maintaining technology across an organisation, you’re often faced with the situation that you would like to deliver much more than is possible with the time, money or team you have. The choices you make to not do certain work have an impact, and must be made carefully.

The most common response to this situation is prioritisation — weighing up the relative importance of work according to factors such as feasibility, complexity, return on investment and benefits to users, or even attempting to align everything to lofty cross-organisational goals.

Prioritisation using these factors becomes very challenging when comparing diverse potential investments: a new digital service vs. upgrading laptops vs. paying down technical debt vs. making systems more compliant. This challenge can be reduced by placing more emphasis on balance over prioritisation of portfolios.

In this post, I use the term ‘portfolio’ to mean a managed collection of technology-related activities an organisation undertakes within fixed resource constraints

The importance of balancing

Portfolio balancing is the idea that leaders and teams should consciously choose the balance of their portfolio between different categories of work, before they prioritise it.

It is part of the thinking applied to portfolio, project and programme management and seen in agile methodologies like Kanban as ‘classes of service’. However these are rarely applied effectively at portfolio level, where balance has the largest impact.

I’d like to argue that the balance of a portfolio is its most important characteristic, and being more conscious of it will help determine success over the long term, and keep technology sustainably healthy and effective. And conversely, that the absence of balancing, and simplistic prioritisation is a likely key root cause behind the accumulation of toxic legacy technology.

Categorisation

Categories can provide insight into the interdependence factors, and the long term impacts of this, and allow for more strategic decision-making.

Below are some categories to help illustrate portfolio balance (adapt to your context as necessary). Most work will not fall cleanly into any one category, so categorisation should be done according to the primary purpose of the work.

1. Creating

Creating or changing technology to gain new value for the organisation or its users.

e.g. building a new product

2. Maintaining

Maintaining the existing technology. In its most basic form, it supports the continued existence of the technology.

e.g. paying the hosting bill each month, renewing a certificate, fixing minor bugs or responding to minor incidents

3. Renewing

Going beyond just maintenance, to ensure the long term health of technology. This may involve refactoring, re-architecting, re-platforming, decommissioning, combining/disaggregating, moving between bespoke and commodity solutions when markets change, and/or decommissioning.

So often, this is the missing category of work.

e.g. upgrading a major component, like a database or moving from private datacentres to public cloud

4. Enabling

Delivering services, platforms or tools which provide value to those delivering. Enabling work does not provide value directly to your service users.

e.g. building a standard payments API or building a hosting platform

5. Reacting

Managing incidents which occur, including remediation. Also, responding to the ending of commercial contracts with extensions or like-for-like supplier changes.

e.g. handling a major incident, including investigation and remediation or changing managed hosting supplier with like-for-like services

The balancing process

Portfolio balancing will require new data — the portfolio items must be categorised. Whilst items may not fit comfortably into single categories, it is useful to characterise them, and to do so according to their primary purpose.

Once you have this data, the first decision to make is about whether to change the balance. Should we do more creating? Are we not doing enough renewing? The output is the same — an expression of balance between categories. For example this could be through percentage budget assignment, team allocation, or something more flexible.

e.g. for the financial year 20/21 we will ring-fence 10% of the budget to work categorised as enabling

Once you have balanced the portfolio, the next consideration should be prioritisation. Within each category, conventional prioritisation should happen. Once this has taken place, it may expose some stark issues — perhaps an exciting new project is no longer affordable? Perhaps there isn’t enough money to upgrade core systems? The hard job of deciding what a team or organisation should do can then happen around two factors: both balance and prioritisation.

Why is portfolio balancing needed?

1. Better strategic decision making

Balancing is important to be able to make more conscious decisions about what to deliver with constrained resources.

2. Better financial planning

Accountancy has its own categories, in particular capital and operational expenditure. These categories hide complexity and interdependence, and encourage a very static view of a technology estate.

Using categories similar to those illustrated above will facilitate better financial planning. They show how investment in different categories have an impact on others. These can still be translated into standard accountancy practice, but sharing and explaining these categories to finance professional working with technologists will improve communication and mutual understanding of the budget.

3. Better risk management

Risk in technology at the scale of large organisations becomes too complex to work through prioritisation alone. You can’t simply play whack-a-mole with impact-likelihood scores, prioritise controls and mitigations, and hope that risk reduces overall.

Categories allow leaders to make conscious decisions about investment in risk reduction. These decisions can be communicated to audit and risk functions, facilitating better conversations about the interaction of risk and delivery.

4. Better cyber security

Although just one type of risk, it’s worth calling out cyber security in the context of balanced portfolios. Too many organisations are caught in the trap where ‘best practice’ or ‘basic cyber hygiene’ (patching, upgrading, adherence to standards) has become seemingly unaffordable due to the volume of legacy technology.

I believe this is principally because basic cyber hygiene is routinely deprioritised. Upgrading a web server just doesn’t seem as important as Business Priority X in a straight comparison. Creating, and ring-fencing investment in categories can create space for good cyber practice.

5. Better understanding of interdependence

Categories are not exclusive — they interact dynamically. Prioritisation often neglects this, or reduces it to a more simplistic idea of project or technical dependency.

For example:

• Creating a lot of technology will increase pressure to maintain and renew
• If you do less renewing, it increases demand for reacting
• If you do more enabling work, throughput for everything else can increase
• If you ‘just focus on the basics’ such as maintaining and reacting, you’ll stop creating and the organisation won’t be able to change or adapt

6. Help identify neglected categories

A good principle to begin with, is to never reduce any category to zero. Prioritisation often encourages this by prioritising work with more measurable outcomes aligned directly to business goals (which sounds superficially like the right thing to do).

Categories which take the organisation forward such as creating and enabling are the work which drives the culture and morale of your organisation. If these are not given time, funding and skills to deliver, the organisation will lose skilled and knowledgeable people.

Categories which we seek to avoid, should be given the time, funding and skills they need to deliver. No organisation can build technology well enough to avoid incidents, so reacting must be a considered part of the portfolio balance.

The skills your organisation needs for each category will vary — designers spend more time creating, site reliability engineers spend more time maintaining and reacting. Portfolio balance allows more conscious thinking about the connections between the work being done, and the organisation structure you have.

Credits

I have been working closely with my colleague Helen Mott @mottlehen on many of these ideas, so a big thank you to her! Thanks to @pgchamberlin for suggesting the term ‘renewal’ for the commonly missing category.