Random Acts of Architecture

Developing and improving software or IT services is challenging. With change the only constant and competition fierce, organizations often live and die by their ability to adapt.

Unfortunately, most organizations are not structured to adapt quickly. Many fool themselves with superficial agile practices without building any real understanding. Some hire external consultants whose benefit ends with their consulting engagement. Others undergo discrete transformations, unaware that adaptation is continuous.

I find it helpful to think of software or IT services requiring three primary foci.

The first is automation. Usually fulfilled by software developers, this is what most people think of with software or IT services.

The second is knowing the customer. For example, an IT service targeting health may need to understand how doctors or clinicians work. A network management service requires understanding routers, networking and network security.

Multi-sided markets may require understanding multiple customer types. Uber, for example, requires understanding both passengers and drivers.

The third is providing the service. This focus includes onboarding/offboarding, billing, service desk/support, security and scalability/performance. Multiple roles within an organization often handle these.

The challenge with most organizations, and a significant reason they have difficulty adapting, is they hire specialists for each focus. Software developers, for example, are usually hired solely for their ability to write software.

Unfortunately, understanding one focus can only take you so far. For example, if you only understand the customer, you risk creating an uneconomical or impractical service. If you only understand automation, you risk automating things of little value. If you do not understand delivering a service, you will struggle to provide one efficiently and economically.

There are exceptions. Product Management is often familiar with the customer (second focus) and providing a service (third). They understand the economics and finances. However, they often distance themselves from automation to prevent a bias toward what their existing services can easily provide.

Management also absorbs a little of all three foci. However, their understanding is often shallow. Their attention is usually broader than just product improvement.

The solution is to identify and then incentivize specific staff to generalize across all three foci. They need to learn enough of each to (1) understand the impacts of a decision or change in one focus on the other foci and (2) identify when opportunities in one can provide a nett benefit.

By comparison, separate individuals struggle with this, even those who communicate well. We often think we know more about others’ roles and needs than we actually do. There is a “critical mass” of understanding that these generalists need to achieve.

The concept of hiring generalists or “T-shaped” skills is not new. However, this usually applies within one of the foci. For example, software developer generalists are often “full stack” developers, not ones familiar with non-technical areas.

Finding generalists is difficult. Each business is different and has different needs. While many universities offer combined IT and business degrees, they usually produce IT management candidates. While some universities provide mixed degrees, they often neglect the third focus on effective service delivery.

Similarly, many want to specialize and master their chosen craft. It is a better understood and measured career trajectory.

Importantly, these generalists do not replace specialists. Organizations need sufficient specialists to deliver each focus.

However, the best generalist candidates demonstrate interest or passion across all three areas. Submerge them in each focus, challenge them to translate the challenges and benefits of each and then task them with identifying bottlenecks and potential improvements.

One of the first hurdles they usually encounter is management structures. Organizations are usually built around similar skills, not business output. Organizations usually incentivize teams or departments around cost minimization and throughput, not customer value.

However, with executive support, creating and supporting these generalists will break down barriers and help the organization adapt quickly. Without doing so, organizations will stagnate, often too siloed or distracted by short-term problems.

Charles Darwin was right. Natural selection applies to organizations, too. Those that cannot adapt will be replaced by those that can.