Why I Tell My Clients to Stop Hiring and Start Automating

The hiring reflex

Every growing business hits the same wall. There is too much work and not enough people. The natural response is to hire someone. Put up a job ad, interview a few candidates, bring someone on board.

I used to do this too. And every time, I ran into the same problems. It took months to find the right person. Training took longer than expected. And within a few weeks, the new hire was drowning in the same mess that everyone else was drowning in, because the problem was never a people shortage. It was a systems shortage.

Now, when a client tells me they need to hire, my first question is always: "What would that person actually do all day?" The answer usually reveals that 60-70% of the role is repetitive work that should not require a human at all.

The real cost of a hire

People think about salary. Maybe benefits. Perhaps recruitment fees. But the real cost of a hire goes much further:

• Recruitment time: 20-40 hours of your time writing the ad, reviewing CVs, interviewing candidates
• Onboarding: 2-4 weeks before they are productive
• Management overhead: Ongoing supervision, reviews, one-to-ones
• Risk: If it does not work out, you are back to square one in 3-6 months, minus the salary you paid and the time you invested

For a 30k salary role, the true first-year cost is closer to 40-45k once you factor everything in. And you still have a single point of failure. If that person is ill, on holiday, or leaves, the work stops.

Compare that to an automation that costs a few hundred pounds to build, runs 24/7, never calls in sick, and scales instantly.

What should be automated

Not everything can or should be automated. But a surprising amount can. Here is how I categorise it:

Automate immediately:

• Data entry between systems (copying from one tool to another)
• Standard email responses (confirmations, follow-ups, reminders)
• Report generation (pulling data and formatting it)
• Social media scheduling (once content is approved)
• Invoice creation and payment reminders
• Meeting scheduling and calendar management

Automate with human review:

• Client communication (AI drafts, human sends)
• Content creation (AI assists, human edits)
• Lead qualification (system scores, human decides)
• Quality checks (automated flags, human resolves)

Keep human:

• Strategic decisions
• Relationship building
• Creative direction
• Complex problem solving
• Anything requiring empathy or nuance

A real example

One of my clients, a recruitment agency with 8 staff, came to me wanting to hire two more consultants. They were struggling to keep up with candidate sourcing, client communication, and admin.

Instead of hiring, we spent four weeks building automation:

1. CV parsing: New applications get automatically sorted by role, experience level, and location. Before this, a junior staff member spent 3 hours a day doing it manually.

1. Client updates: Weekly progress reports that used to take 2 hours per client now generate automatically from the CRM data. A consultant reviews them for 5 minutes before they send.

1. Interview scheduling: Candidates now book their own interview slots through a system that checks consultant availability automatically. No more email ping-pong.

1. Follow-up sequences: After a placement, automated check-ins go out at 1 week, 1 month, and 3 months. These used to fall through the cracks entirely.

The result? They did not hire those two consultants. The existing team got 15+ hours back per week, collectively. They used that time to build relationships and close more deals. Revenue went up 20% in the next quarter with the same headcount.

The objections I always hear

"Our work is too complex to automate." Some of it is. Most of it is not. The complex work is probably only 30-40% of the total. The rest is admin, communication, and data handling.

"Our clients expect a personal touch." They do. And they are not getting it right now because your team is buried in admin. Automate the admin and your team has more time for the personal stuff.

"We are not a tech company." You do not need to be. The tools available now, like n8n, Zapier, and Make, have visual interfaces. Your operations manager can learn them in a week.

"What if the automation breaks?" It will sometimes. But when a human makes errors due to fatigue or overwork, nobody calls it "breaking." Set up monitoring, have a fallback, and fix issues as they come. The error rate will still be lower than a stressed employee rushing through repetitive tasks.

The right order of operations

1. Document what everyone actually does (time-tracking for 2 weeks)
2. Identify the repetitive, rule-based tasks
3. Build automations for the top 5 time-wasters
4. Measure the time saved
5. Reassign that time to high-value work
6. Only then, if there is still more work than people, hire

When you do eventually hire, you will be hiring for a much better role. Instead of "do this admin all day," it is "use your skills and judgement to grow this part of the business." That attracts better candidates and they stay longer.

The best businesses I work with treat automation as infrastructure, not a nice-to-have. It is the foundation that makes everything else possible.