Otter Networks logo

By Andrew Holway - Otter Networks Founder - 12th February 2019

Pairing is the New Training: Why Development Teams Need to Rethink Traditional Methods of Knowledge Sharing

Training is letting us down. It’s time for a new approach.

Pairing vs. Training

Research suggests that we retain just 30% of what we see and hear in a lecture or demonstration. In contrast, we remember a whopping 90% of what we learn when we use knowledge immediately or teach it to someone else.

If this is true, then why do most training models still consist of lecturing and demonstrating?

As technology grows increasingly more powerful and developers gain opportunities to take on new roles, it’s becoming imperative for development teams to adopt more effective knowledge sharing techniques.

Training, as we know it, just isn’t cutting it. It probably never did – but in 2019, its shortcomings are more apparent than ever. Now it’s time to take a long, hard look at how we can ditch training as a primary method of knowledge sharing and replace it with a better model: pairing.

But first, let’s dig into why training is letting us down so miserably.

Accepting reality: How training fails development teams

Training – the most commonly used method of knowledge exchange in the workplace – rarely allows trainees to get into the driver’s seat. In a typical training scenario, trainees don’t get the chance to make meaningful mistakes. A healthy amount of slipping up or flailing around is essential for retention of information in general, but it’s even more important when navigating new technologies and infrastructure.

If we do something ourselves instead of just hear about it, we’re able to make mistakes, correct them and therefore absorb more information through increased concentration. This helps to combat what’s known as the Forgetting Curve – a typical graph that shows how information is lost over time when there are no attempts to retain it.

Without active repetition of learnings, we tend to forget half of newly learned knowledge in a matter of days.

Forgetting Curve

By its very nature, training perpetuates this barrier to knowledge retention. It doesn’t allow for learning by doing, nor repetition of information. Simply put: it doesn’t allow for the acquisition of tacit knowledge.

Learning by doing: The crucial role of tacit knowledge

Experts at a particular task are often said to possess tacit knowledge – or knowledge that cannot easily be articulated or communicated. Tacit knowledge is a skill set gained through repeated activity, extensive personal content and high quality social interaction. One might think of it as knowledge that’s “captured” within a certain context or community.

Training isn’t successful in exchanging tacit knowledge because it doesn’t emphasize a learning-by-doing approach. Think about it: trainers often teach by doing things themselves. They build artifacts – or by-products of their teaching process – in order to demonstrate a concept. But these artifacts have no real world value to the trainees; they’re simply an example. Because trainees cannot interact further with these artifacts once training is completed, repetition of specific learned skills is not possible. The real value is lost without ownership of artifacts. The Forgetting Curve begins its cruel trajectory.

Finding the ideal: Why pairing is the new standard of knowledge transfer

What development teams need is a method for knowledge exchange that builds tacit knowledge, and therefore gives the learners real ownership and retention of their skills.

Pairing is the value-driven educational approach that modern development teams need to innovate and succeed. Derived from the effective practice of pair programming in which developers – each with separate roles – team up to code using one machine, pairing brings the learner into a position of control from the beginning. With the guidance and knowledge of a teaching Observer, the learner becomes the Driver and absorbs knowledge by building on their own. The learner is able to achieve successes and make mistakes in a controlled and nurturing environment.

The repetition associated with this approach transforms the Forgetting Curve into something wholly different.

Overcoming Forgetting Curve

Let’s look at an example. At Otter Networks, we work closely with CTOs and engineers who are looking to transform their company’s infrastructure and deploy faster. Our approach to making this happen for them is achieved in a one-week Skills Sprint, during which we pair with developers while they build a flexible CI/CD infrastructure using Kubernetes. Not only is the team left with valuable knowledge about how to manage brand new infrastructure – but they enjoy true ownership of it because they’ve built it themselves.

Pairing for the future of development

The demand for developers is only increasing, and it’s time to get real about how we’re educating each other and providing value with available tools and skill sets. Pairing’s learning by doing structure makes it possible for the learner to build expert knowledge – and therefore ownership – of any task and associated artifacts.

It also inherently promotes the ideas of autonomy and empowerment – something vital with all the powerful tooling capabilities on the market. Developers can now control their skills to the fullest extent and product valuable results, with zero reliance on project managers or DevOps engineers.

We’ve seen the results of pairing through our own experiences in the design sprints, and they’re impressive. Knowledge exchange through pairing allows dev teams to be nimble, quick and innovative; it allows them to more easily test potentially fruitful ideas, validate theories or meet major product deadlines. This is a method of education that can match the rate of transformation within our industry.

Want to know more? Drop us a line. We regularly host webinars and events about knowledge exchange, productivity engineering and Kubernetes.