Skip to main content
Back to all posts

business automation

What to Automate First in Your Small Business

Most automation guides hand you a tool list. Here is a simple test for picking your first task, what to leave alone, and how to check it worked.

Charlie Johns working from a laptop with a sea view, planning which small business tasks to automate first

Search for what to automate first and you will find a list. Invoicing, data entry, social posts, email sequences, a dozen tools to try. It looks helpful until you realise it answers a different question from the one you asked. You did not want a list of everything that could be automated. You wanted to know where to start so that the first attempt actually pays off.

Here is the short answer. Automate the task that costs you money when it is slow or forgotten, before the one that is simply annoying. The rest of this post explains how to tell the difference, what to leave well alone, and how to check that it worked.

Why most advice sends you to the wrong place

The usual advice is to automate whatever you repeat most often. It is not wrong, but it ranks tasks by how irritating they are, not by what they cost you. Re-typing the same details into a spreadsheet is irritating. Losing a customer because nobody replied for six hours is expensive. Those are not the same problem, and the second one rarely makes the top of a “time saving” list.

The appetite is certainly there. Salesforce’s small and medium business trends report, which surveyed 3,350 small business leaders worldwide, found 75% of them investing in AI. And Xero’s white paper on AI for small business, covered in its guide to the small business time tax, reports daily AI users were more than twice as likely to say revenue had risen, 28% against 12%. Treat that second figure with some care. It is self-reported, and businesses that are already well run may simply be the ones adopting tools sooner. Still, the direction is clear, and the businesses doing well with automation are not doing it at random.

The three-question test

Before you touch any software, take the task you are considering and ask three things.

First, does it happen at least weekly? A task that crops up twice a year is not worth the setup time. Second, is the rule the same every time? If you can write down “when X happens, do Y” without adding “but it depends”, a machine can follow it. Third, does delay or forgetfulness cost you money? This is the question most guides skip, and it is the one that decides what goes first.

A task that scores yes on all three is a strong candidate. A task that only scores yes on the first two is a tidy-up for later. If you want a feel for how the third question plays out in practice, what a missed call actually costs your business shows how to put a figure on it.

Start with the leaks

For most owner-run service businesses, the first automations belong in the gaps between a customer showing interest and a customer paying you. That is where work quietly disappears, and where speed matters most.

Missed calls and new enquiries. If you are on a roof, in a treatment room or under a sink, you cannot answer every call. An instant text reply that says you will ring back, and asks what they need, keeps the conversation alive. The setup behind answering every call is simple, and the effect on lost enquiries is usually bigger than anything else you could automate this month.

Quote follow-up. Most quotes are not rejected. They are forgotten, by you and by the customer. A couple of polite, well-timed nudges win back work you had already written off. The approach in texts that win back work covers what to say and when to stop.

Review requests. Asking at the end of every finished job, while the customer is still pleased, brings in far more reviews than remembering to ask when you have a spare moment. Reviews feed straight back into how easily the next customer finds and trusts you.

These three are the heart of what I built into No Lead Left Behind, and not by accident. They scored highest on all three questions, again and again, in businesses that came to me wanting something quite different.

The second tier: weekly chores worth handing over

Once the leaks are plugged, move on to the repetitive tasks that eat your evenings. Invoice reminders that go out on a schedule, so you stop chasing payments personally. Booking confirmations and appointment reminders, which cut no-shows. A welcome message that goes to every new customer with the same useful information. Weekly reports that arrive in your inbox rather than being built by hand.

These are all good candidates, and all easy to measure. They are simply not urgent in the way an unanswered enquiry is. If you are also thinking about outbound messages to a list, email marketing for a small business is a sensible place to apply the same thinking.

What to leave alone, at least for now

Some work should stay in human hands, and not just because it is hard to automate. It is because the customer can tell.

Pricing a complicated job needs judgement. Handling a complaint needs empathy and the authority to put things right. Delivering bad news, negotiating, and anything involving a long-standing customer relationship all belong with you. A useful rule is to automate the first reply and the routine follow-up, then hand over to a real person the moment the conversation needs thought.

It is also worth pausing before you automate anything that touches personal data. A reply to someone who has just contacted you is quite different from a marketing message to someone who has not asked for one, and the rules on consent differ from country to country. Check what applies where you operate, and keep a simple way for people to opt out.

How to know it worked

This is the part almost nobody does. Before you switch anything on, write down three numbers: how many enquiries you receive in a typical month, how many you answer within an hour, and how many become paying jobs. It takes ten minutes.

After 30 days, look again. If more enquiries are being answered quickly and more are turning into work, it is paying for itself. If nothing has moved, change the message before you change the tool, because the wording is usually the weak point rather than the software. The same discipline applies to any marketing spend, and how to measure marketing ROI walks through it step by step.

A realistic first month

Week one, pick one task using the three questions and write down your baseline numbers. Week two, write the message or rule in plain English, as if explaining it to a new member of staff. Week three, switch it on and read what customers actually receive for the first few days. Week four, review the numbers and decide whether to add a second automation or fix the first.

Resist the urge to automate five things at once. One that works is worth far more than five that half work, and you will learn what your customers respond to before you build anything bigger.

If you are weighing up who should set this up, whether that is you, a freelancer or a larger team, choosing between a marketing agency and a freelancer covers the questions worth asking before you hire anyone.

Where to go from here

You do not need a big project. You need to know which one task is quietly costing you the most, and to fix that first.

If you would like help finding it, the free Digital Blind Spot Report shows where enquiries and customers are slipping through your current setup, with no commitment. You can also read more about how I work, see what my marketing packages include, or book a 20-minute call and I will tell you plainly which task I would automate first in your business.

Common Questions

Things people ask about this.

What should a small business automate first?

Start with whatever costs you money when it is slow or forgotten, not whatever is merely tedious. For most owner-run businesses that means replying to missed calls and new enquiries, following up quotes, and asking for reviews after a job. Invoicing reminders and booking confirmations are good second steps. Pick one, run it for a month, and only then add another.

How much does it cost to automate small business tasks?

It varies widely, from a few pounds a month for a simple tool you set up yourself to a monthly fee for someone to build and maintain it for you. The better question is what the task costs you now. If a missed enquiry is worth a few hundred pounds and you lose several a month, even a modest tool pays for itself quickly. Work out that number before you compare prices.

Which tasks should I never automate?

Anything that needs judgement or carries emotional weight. Pricing a complicated job, handling a complaint, delivering bad news and negotiating a contract all go better with a person. A good rule is to automate the first reply and the routine follow-up, then hand over to a human as soon as the conversation needs real thought.

Do I need to be technical to automate my business?

No. Most first automations, such as text replies, review requests and invoice reminders, run on tools built for non-technical owners. The harder part is deciding what the rule should be and what the message should say, which is a business question rather than a technical one. If you would rather not set it up yourself, that is a sensible thing to hand to someone else.

How do I know if my automation is working?

Measure before you switch it on. Write down how many enquiries you get, how many you reply to within an hour, and how many turn into jobs. After 30 days, compare. If the number of enquiries answered quickly has gone up and nobody has complained about the messages, it is working. If the numbers have not moved, change the message before you change the tool.

Want the same thinking applied to your business?

Let's map out your next 90 days.

Book a 20-minute call. I will listen to what is going on, ask the questions that matter, and tell you honestly whether I can help. No pitch deck. No pressure.

Book a 20-minute call See the fix menu Or get the free audit