Photo: Alesia Kazantceva

Why some quit, and some stay: a surprising take

The “intention to quit” surveys aren’t a great predictor of employee attrition — something else is. And it has to do with networks of knowledge.

The engagement gap. The new implicit contract. YOLO. All have a bearing on the “Great Resignation”. But here’s a different perspective that leaders should urgently consider — with some very practical implications for the retention of senior professionals.

For some time, some companies have been using enterprise social network analysis to identify the reason for employee attrition. The foundational social-network science on this is established (MIT’s Honest Signals and Social Physics, for instance). We know that if we want, we can often anticipate and possibly prevent people’s resignations by looking at their network signature. (Examples of the research here and here.)

But is there a more proactive picture? Is there a “prevention-medicine” equivalent, instead of just curing the individual problem when it happens? There’s ground to think so.

Two reasons why good people go

There are many reasons why people leave but many of them relate to these two questions:

  1. Do they feel they’re part of a (good) group? Do their strong (e.g. manager) and weak (colleagues met at the watercooler) network ties give them a sense of emotional support, group “flow”, as well as functional help?
  2. Are they able to make an impact — a recognized and rewarded one — where they are? People’s impact depends on which depends on being associated with the right things to do, getting access to knowledge and learning, doing work frictionlessly, and getting access to the right people to influence actions. The last one is especially important for senior people who typically work across organizational silos, and outside of fixed, rule-based workflows.

Interestingly, there’s evidence that the answer to both questions depends on network structure, but let’s focus on the “ability to make an impact” because that’s where many of the surprises are.

Having impact through a network means that, at parity of individual capabilities, (a) you can access hard-to-find (e.g. undocumented or unindexed) information and you can do that because your direct and indirect networks have it and (b) armed with the resulting recommendations, you can influence how things are done by nudging directly and indirectly the right stakeholders at the right time. Specifically, getting (b) done in the case of unpopular or counterintuitive decisions is what makes companies strong, and managers successful.

(The spreading of useful but against-the-grain knowledge has been researched thoroughly: some call it “complex contagion”: it is information that, unlike easy “memes”, isn’t believed and amplified by the network unless the recipient hears it from multiple, unrelated sources. See here for instance).

The tipping point

Good people who fail to create a strong network in their current company often can’t make enough impact, and as a consequence, they may have less to lose and could end up leaving. For them, the skills that they have — not the network and the related social capital — are the biggest asset. Those skills are marketable externally, and the loss of a company network doesn’t penalize them much.

Think for instance about people whose networks are particularly strong outside of the company (e.g. some salespeople, senior consultants): they would also have less to lose when changing employers. Also, research using LinkedIn data over five years, completed in 2022, shows that a large number of weak ties can help people find new jobs. That’s especially true when remote work complements physical-office work, which makes moving elsewhere easier.

Contrast this situation with that of good people who have been in a company for long: their network’s ability to generate impact represents a higher proportion of their professional assets. And it’s not as portable: that is, they would lose a lot of that if they moved elsewhere.

That’s what the simplified chart below illustrates. The hypothetical at-risk employee is the blue, larger icon in the left quadrant.

The real brain in the brain drain

Step back for a second, because there’s an even bigger picture. What we are saying is that people use a “collective brain” to sense, remember, create, decide and learn. That’s what generates the real impact.

It is a form of what some call a supermind: a cognitive engine constituted partially by that individual person’s brain, but also by the “neural” network structure in their organization, or in their ecosystem.

That brain is marketable internally, or externally — depending on where the network is comparatively strongest.

What can be done

A clear implication: in companies where knowledge is easy to retrieve, and where networks are easy to access, relatively new people typically become impactful faster, which should reduce the probability of their attrition.

Conversely, companies where the network is all-important yet tribal and knowledge isn’t easy to retrieve will have an easier time retaining people who have been there for a long time: these employees’ success is predicated on painstakingly building that network, which has now become the primary asset, and possibly a competitive moat against other employees.

So while the creation of internal networks is important for long-term retention, it can become a hurdle for others, especially but not exclusively new people, and that will make them attrite. But if companies invest in enabling people’s “network-based impact” and improve knowledge access (including learning and knowledge management), the collaboration tools (e.g. virtual whiteboards, asynchronous conversations), and the ability to network effectively (including serendipitous encounters, affinity groups), they might be able to retain more of their best employees. Individual managers are responsible to make some of this happen, and they may need specific sensitization and training. But CEOs need to invest in augmenting the digital infrastructure that caters to the broader system and making the related change management happen.

There’s a clear choice to be made, right there.

Have a look at these previous blogs (here, here, here, and here) for more.

This post complements the tech-driven organizational design materials at and my previous blog posts (here, here, and here, for example) on designing an AI-augmented collective intelligence. Read them if you’re interested in using these techniques in your own organization, and get in touch if you want to discuss.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store


Chief Innovation Officer at Genpact, Head of Innovation Design at MIT Collective Intelligence Design Lab, advisory boards