Beating the IT brain drain: How to break the cycle of vanishing skills

How much do you know about your own in-house knowledge base? How do you stop that knowledge draining away when people leave? What about the areas in which you have expertise that you didn't know existed in the first place?

Twitter University founder Marko Gargenta tells IT Pro exposing those "unknown knowns" – as late US defense secretary Donald Rumsfeld might have said – can be critical. “There's tacit knowledge and there are documented nuggets of wisdom, like policies, procedures, things that are stable," says Gargenta, currently chief executive of provider PlusPlus.

Within tech companies, especially those which continually innovate and reinvent themselves, a lot of knowledge falls into the less explicit category, Gargenta points out. Often firms rely on the ad hoc tap on the shoulder in situ to transfer tacit information, but this isn't efficient.

Instead, teams should discover tacit knowledge – for example by interviewing both up and down the hierarchy, asking lots of questions, then triangulating to expose knowledge gaps. They should also find challenges and develop multiple perspectives. Once made explicit, that knowledge can be documented and taught.

"It's usually like some common tool or library that a lot of people need to use and it's poorly understood," Gargenta says. "Then we can come up with a solution. Is it a workshop? A tech talk? A document or mentorship?"

Keeping knowledge alive

Live learning through hands-on workshops that generate interactivity and engagement can be cost-effective in many cases – not only transferring knowledge but ensuring people can adapt and use that knowledge in future. A regular hour or two, requiring relatively little preparation on the participants' parts, can often do the job.

Annee Bayeux, chief learning strategist at Degreed, suggests looking at staff retention too – it can't all be about offboarding. Also, when people leave, it can be partly because knowledge isn't being fully used and transferred across the company; growth potentials aren't being fulfilled.

Government figures suggest 46% of UK businesses struggled through 2019-2021 to recruit for roles that require data skills, defined broadly.

"From day one, make sure you're developing them – and their managers, including how to coach, mentor, and give feedback," Bayeux says. "Even once they reach peak performance, people want to be learning new skills to stay relevant, or go deeper or wider."

Have you got people in accounts that could move into data science, or vice versa? If staff have 50% of a needed skillset, why not upskill them on the other 50% to give them more professional mobility? Overlooked opportunities can lead to staff becoming stale or disgruntled and leaving – taking their knowledge with them.

Capture knowledge from day one, and also look to knowledge development and delivery according to individual desire and requirement, Bayeux emphasises.

Caroline Carruthers, chief executive officer at global data consultancy Carruthers and Jackson, warns data isn't synonymous with information or knowledge either, so don't simply rely on a digital solution. Organisations must first clearly understand what's important to them. This can even mean discussing the nuance of key concepts, right down to defining the customer. Agreement on these things is "a lot less common than people think" yet can be the "crown jewel" of business success.

"Ask questions, be curious – that's how you get to the key assets," Carruthers says. "Look at purpose, people, methods and tools."

Focus on what matters most

Cristina Querzè, HR director – sales and Western Europe at Vertiv, says her company created three different academies to develop and retain technical knowledge in identified key competencies. Although organisational culture is also crucial, Vertiv's key asset is its technical expertise. "We ask what our teams need to know, and start with that," Querzè says. "Then there are processes to codify that."

Every six months, lists of critical knowledge for, say, sales engineers will be reviewed. Technical committees also work to identify subject matter experts and authors who can contribute to digestible lessons or content packages.

Mentoring and reverse-mentoring is useful. Learning in the flow of work is critical; it's not about creating an encyclopaedia of Vertiv knowledge, she emphasises.

"Industry has changed, customers have changed, knowledge needs to be continuously updated," Querzè says. "70% of the knowledge comes from experience and practice; 20% usually comes from serving others or the like; only 10% will be about formal training courses."

Jaco Vermeulen, chief technology officer at BML Digital, says many companies see it as about transferring knowledge from one brain to another and just don't do enough systematisation. Organisations need to move beyond simply avoiding, for instance, "reverse engineering a massive Excel spreadsheet" when someone leaves.

"Fragmented or very federated organisations need to be brought together to start to operate in harmony across the board," Vermeulen adds.

When systemisation is tied to a specific business outcome, do also communicate that it's not about removing people from the equation. "People are mostly eager to share what they know – but that's been a failing of business leadership and focus, so do that work today," Vermeulen warns. "Then IP becomes configurable rather than static."

Teresa Dietrich, chief product and technology officer at Stack Overflow, adds that it's crucial to get on top of this requirement, with knowledge soon becoming outdated and tech seeing high staff turnover. “Knowledge must be shared openly throughout their tenure and long before a departure," Dietrich advises.