What Should We Automate?

AI is not here to replace recruiters. It’s here to make us faster. The real question is not whether to use it, but what exactly should we automate?

At its best, AI is a data-processing engine. Large language models (LLMs) can take messy, unstructured information and make it clear in seconds. For recruiters, that means turning mountains of profiles, CVs, and market data into insights we can actually use. The trick is knowing which parts of the process should be handed over to the machine, and which parts only a recruiter can truly handle.

Spotting What Others Miss

A rookie recruiter might skim a LinkedIn profile and miss a key achievement buried in the text. An LLM can pull those accomplishments out and present them cleanly. Suddenly, what might have been overlooked becomes front and centre. This is particularly useful in markets like Japan or Germany where profiles can be sparse, or where small but significant details get buried in long tenure. AI becomes your second set of eyes, ensuring nothing slips past.

Even better, AI can tag candidates automatically. It can infer whether someone’s showing upward trajectory, consistent tenure, or a career pivot. These are things experienced recruiters do instinctively, but AI gives us a way to scale that wisdom. It’s not perfect, but it’s directionally correct - and in recruitment, that’s often enough to make better, faster decisions. Instead of spending hours combing through profiles, you get a structured overview in seconds that helps you make smarter choices.

Building Long Lists in Seconds

Traditionally, recruiters built long lists one by one. AI can generate one instantly. It takes your criteria, maps them across the data, and hands you a starting point. The key is not to blindly trust it, but to use it as the first cut. The recruiter’s job is still to review, validate, and curate that list into a shortlist of real contenders. You still do your job, but you are much faster at it thanks to AI. And the time you save here can be reinvested into the higher-value work - conversations, negotiations, and market intelligence.

Filtering In, Not Filtering Out

I don’t like to say AI “filters people out.” Instead, it “filters in” the candidates most worth your attention. That subtle shift matters - the recruiter still owns the final call. Good recruiters fill in the blanks, see patterns others miss, and recognise who has the highest probability of success in a role. AI can take hard or inferred data such as consistency of tenure, career trajectory, impact, and so on, to push the top candidates to the top of your lists. It essentially does the heavy lifting of triage, so you can spend your time on the highest-probability outcomes rather than sifting through noise.

Where AI Shouldn’t Be Used

Typing outreach emails for executive search? That’s not a job for AI. If you’re engaging at the C-level, your message needs to be crafted with nuance, context, and care. On the other hand, if you’re running a high-volume campaign, AI can help sort the noise and highlight referral pathways. It can show you who in your network might open the right door - that’s worth its weight in gold. Used well, it becomes less about spam and more about precision targeting, which is exactly what good clients expect.

The Importance of Referrals in an AI World

Here’s the hard truth: AI will dominate high-volume hiring. Corporate recruiters and their RPO suppliers will lean on it heavily, contingency agencies will deploy it when desperate, and in contract staffing it will become the default engine to blast jobs accurately and at speed. But that’s a race to the bottom. The one area traditional contingency agencies can still win is through referrals, and referrals come from trust.

Building Trust

Trust is not built by spamming inboxes - it comes from relationships. And relationships come from showing up with more than a polished suit and good manners. They require knowledge. This is why I used to teach my recruiters to become Subject Matter Experts (SMEs) in their sector. When you know your market inside-out, people trust you. When you can ask sharper questions than the hiring manager, you stand out. And when candidates see you as an industry insider rather than “just another recruiter,” they open their network to you.

As an aside, I once ran a book club at an agency. Everyone had to present how what they read could be applied to their work. One recruiter, let’s call him Taku, was average at best. But he read a book on computer-aided design, and when he later attended a client briefing, he was able to hold a proper technical conversation with the hiring manager. That single exchange won him two exclusive searches that nobody else in the room even had a shot at. Expertise opens doors - sometimes from a book, sometimes from conversations over years of experience - but always because it builds trust.

Beyond Candidates: Mapping Companies

It’s not just about people. AI can analyse which companies are on an upward trajectory, who the key players are, and where the market is heading. That’s insight recruiters can take back to clients, showing we’re not just pushing CVs, but bringing intelligence to the table. This is where recruitment edges into strategy consulting - advising clients on where the market is going, not just reacting to where it is today.

The Human Edge

Recruitment has always been about imperfect data. The best recruiters know how to interpret it, spot the signals, and build relationships on top of it. That last part - the human connection - is where AI stops and people start. Whether you’re introverted or extroverted, recruitment is about trust. And trust isn’t automated. This is why recruiters who embrace AI will actually become more human, not less - because they’ll spend more time where it matters most: with people.

The Hard Truth

AI will absolutely wipe out a large number of recruiters. Not overnight - it’ll happen slowly, like boiling the frog. One day, those recruiters will look around and wonder why they’re not making money anymore. The reality is that their processes will have become outdated, their speed uncompetitive, and their clients will have moved on to faster, sharper alternatives. The answer will be simple: because they were too slow.

In case you haven’t noticed, it’s already happening. See my post on the barbell effect for more.