AI Automation for Professional Services: Where to Start

Here's something I see every week. A law firm with brilliant solicitors spending 45 minutes a day filing emails. An accounting practice where the senior partner - the person clients are paying for - is manually copying data between systems. A recruitment firm where the founder is still doing their own follow-up emails at 9pm.

Smart people doing dumb tasks. That's the problem.

Most firms are good at what they do. Nobody has shown them what to automate first, and what to leave alone.

So let's fix that.

The 30-45 Rule: Your Starting Point

I ask every client the same question: What takes you 30 to 45 minutes every day that you don't like doing?

The annoying, repetitive stuff that eats your day.

That's where you start.

Those 30 to 45 minute tasks are usually:

• Predictable - they follow a pattern, even if you haven't written it down
• Low-risk - if something goes slightly wrong, nobody gets sued
• High-frequency - they happen daily or weekly, so the time savings compound fast

Do the sum for your own firm. A 30-minute daily task across three people is an hour and a half a day, or roughly 390 hours over a working year of 260 days. Multiply that by their hourly rate and you have the business case. That's what I call The Productivity Cost Calculator, and our AI Savings Calculator runs it on your own numbers.

Two Areas Where AI Actually Works

There's a lot of noise about what AI can do. Most of it is hype.

1. Admin and Operations

This is where the biggest time savings hide. Most firms carry a mountain of admin nobody talks about because it's "just how things are done." We call it the admin layer.

• Email filing and categorisation - automatically sorted, renamed, and logged to the right matter or client file
• Document generation - meeting notes turned into draft client letters, file notes or reports, ready for a person to check and sign
• Data entry - information flowing between your CRM, practice management software, and spreadsheets without anyone retyping it

This is the heart of what we build at Blue Seas AI. The admin comes off your desk first.

2. Client Communication

Your clients want to know where things are up to, and your team can't stop work to tell each one. AI prepares the routine updates, and a person checks them before they go:

• Status updates drafted when a milestone is hit, for a person to check and send
• Follow-ups for the documents and answers a client still owes you
• Client onboarding that collects documents and information before the first meeting

Clients feel looked after, and your team gets that time back for the work only they can do.

What This Looks Like in Practice

Catton & Tondelstrand (20-person family law firm, Sunshine Coast): We built two AI systems. One files 4,000+ emails a month to the right matter in Smokeball. The other turns each consultation into the file note and client letter in under 5 minutes, down from 45. The firm is recovering more than $1.3M a year in productivity, and a lawyer reads and signs everything before it leaves the firm. Read the case study.

SEQ Campers (family business): They started with one repetitive job. That first system has grown into competitor tracking, monthly performance reports and a plan to make the whole business AI-first.

LH Sales Consulting (recruitment): A weekly lead-generation workflow that took about ten hours came down to about 75 minutes, 87% of the time saved.

Different industries, same approach. Find the time sink, automate it, measure the result.

The Amplifier Test: Fix Your Processes First

AI amplifies what you're already doing. If your processes are solid, AI amplifies your success. If your processes are chaos, AI amplifies the chaos.

You can't bolt AI onto a broken process and expect magic. It's like learning basic chords then being asked to play Sweet Child O' Mine in front of 100,000 people.

Before we build anything, we run what I call The Amplifier Test. Amplify the process 10x. Does it create value or problems?

If the answer is problems, the process gets fixed first. That's the difference between a system that works and one that becomes another headache.

If I don't see ROI in this, we're not doing it.

Where NOT to Start

Don't start with the biggest, scariest project. I see firms that want to jump straight into AI-powered legal research or fully automated financial reporting. Those opportunities will come. They're complex and high-stakes, and if something goes wrong the consequences are serious.

Think of AI as a fast, tireless assistant on its first day. Clever, wants to help, but you've got to show it what to do. You wouldn't hand it the corporate credit card on day one.

Start small, build trust, then build bigger.

A good first project:

• Takes 30 to 45 minutes and happens at least weekly
• Follows a repeatable pattern
• Low risk if something goes slightly wrong
• Easy to measure before and after

Get a quick win. Let your team see it working. Then they'll come to you asking what else you can automate. That's the moment AI adoption has landed.

If you run a law firm, the same test is worked through job by job in What a Law Firm Should Automate First, and What It Never Should.

The Lipstick Test

If you're researching AI tools, be careful. The market is flooded with shiny distractions.

Before you sign up for any AI tool, ask yourself: Is this actually new, or is it just ChatGPT with lipstick?

A surprising number of "AI-powered" platforms are a wrapper around the same large language model with a nice logo and a monthly subscription. You can often get the same result for a fraction of the cost with a well-designed automation.

Some tools are useful. Don't pay premium prices for something you already have underneath.

Ready to Find Your Starting Point?

Bart Puszko

Founder and Principal of Blue Seas AI. Queensland Government AI mentor. Winner, 2025 Sunshine Coast Business Awards, Advanced Technology. Top 10 finalist, Australian AI Awards 2025 & 2026.