· 3 min read · Jason Youngjin Koh
AI transformation at small scale
For a large company, AI transformation is a project with a steering committee. For a small business it is simpler: find the work that repeats, and hand it to a system that never forgets.
For a small business, AI transformation means handing repetitive work to AI systems so the owner and team spend their time on customers and decisions. The best starting points are the jobs closest to revenue: answering enquiries, following up leads, producing marketing content, watching ad performance and re-engaging past customers. Start with one, prove it, then add the next.
Where does AI pay back first in a small business?
| Area | What AI does | Who benefits |
|---|---|---|
| Enquiries | Answers common questions and books appointments at any hour | Clinics, academies, service businesses |
| Follow up | Sends the right message to every lead and past customer on time | Every business with a lead list |
| Content | Produces images, video edits, captions and ad variations | Brands and anyone running ads |
| Ad management | Watches campaigns and flags what to scale, fix or stop | Anyone spending on Meta or Google |
| Reporting | Pulls numbers from several tools into one weekly summary | Owners who spend hours on spreadsheets |
| Website | Ships page changes and new landing pages in hours | Anyone waiting on a developer |
What should a small business avoid?
- Buying a tool before defining the job it should do.
- Automating customer conversations that need judgement, such as complaints or clinical questions.
- Letting AI publish claims nobody has checked. Advertising law still applies to AI written copy.
- Feeding customer data into tools without knowing where it is stored.
- Trying to change everything at once, so nobody can tell what worked.
How long does it take to see results?
For a single, well defined job such as enquiry replies or lead follow up, weeks. The first version runs beside the person who currently does the job, and stays only where it matches or beats them. Larger changes, such as connecting the website, booking system and ad accounts so AI can act across them, take a few months and are built in stages.
What does an AI native business look like?
Its website, sales tools and admin systems are connected so AI can read and act across them. A new enquiry is answered, logged, followed up and reported without anyone copying it between apps. Marketing content is produced weekly instead of quarterly. The owner reads a summary of what happened and decides what changes. The team is the same size and does far more.
Where should you start this month?
- List the tasks that repeat every week and roughly how many hours each takes.
- Pick the one closest to revenue, usually enquiry handling or follow up.
- Define what a good result looks like, in a number.
- Run AI on it for a month alongside the current process and compare.
The marketing side of this, job by job. What AI agents do in marketing