Most businesses grow in one of three ways: they keep more of the customers they have, they sell more to them, or they reach new ones. Seven moves cover nearly all of it: set one goal with a number, stop losing customers, sell more to existing ones, take back hours with automation, open a new channel, find partners, and keep watching competitors. Software helps with most of them, but only when you can show in advance that it pays for itself. Below is how to pick a move and run that check.
| Move | What it changes | Where software helps | First number to track |
|---|---|---|---|
| 1. One goal with a number | Focus | A simple dashboard | The goal itself, weekly |
| 2. Keep the customers you have | Revenue you already earned | Reminders, account pages, support tools | Share of customers who buy again |
| 3. Sell more to them | Revenue per customer | Bundles, upgrades, a customer portal | Average spend per customer |
| 4. Take back hours | Cost and capacity | Automation and AI on repeat work | Hours a week on that task |
| 5. Open a new channel | Reach | An online store, an app, a booking page | Sales through the new channel |
| 6. Partner up | Reach at lower cost | Integrations with partners’ tools | Customers who came through partners |
| 7. Watch competitors | What to fix next | Review and price tracking | Reasons customers give for leaving |
“Grow” isn’t a goal. “Add 40 regular customers by the end of March” is. The SMART checklist (specific, measurable, achievable, relevant, time-bound) is old but still useful. Pick one goal per quarter, write it where the team sees it, and check it every week. Everything below is a way to reach it.
You’ve already paid to win the customers you have. Before spending on new ones, find out how many current customers buy again and why the others stop. Ask the ones who left. Often the fix is small: a reminder, a faster reply, an easier way to reorder. Our post on customer obsession covers how to listen in a structured way.
Look at what your customers buy elsewhere that sits next to what they buy from you. Package it. A service business can add a maintenance plan; a shop can add a subscription for things people reorder. Customers who already trust you are usually the cheapest people to sell to.
Many businesses lose hours every week to copying data, formatting documents and answering the same questions. That repeat work is the easiest place to start with automation.
Most firms haven’t started yet:
People are already using it at work. In a Census Bureau survey from March 2026, about 56% of US workers said they used AI for at least one of 11 tasks. Of those who used it in the previous week, 31% said it saved them one to two hours, 15% three to four hours and 15% more than four.
For the recruitment firm Hays, we automated CV formatting. One CV went from up to 2 hours of work to 2 minutes, for 200+ recruiters. Our process automation page covers documents, data entry, reports and access control. A pilot costs $5,000 to $15,000 and takes 2 to 4 weeks.
If customers can only buy by phone or in person, an online channel is the natural next step. In the US, e-commerce made up 17.1% of all retail sales in the second quarter of 2026 and grew 12.2% on the year before, according to the Census Bureau. Start with the smallest version that takes an order or a booking. Grow it once people use it.
Other businesses sell to the same people you do. A gym, a nutrition shop and a sportswear store share customers; so do an accountant, a payroll tool and a bank. Agree on referrals or joint offers, and if your software can connect to theirs, customers will find you inside tools they already use.
When customers leave, ask where they went and why. Read your competitors’ reviews for the complaints they don’t fix. Track their prices once a quarter. This tells you what to improve next, which is better input than guessing.
Before you build anything for growth, do the math on one page:
If the answer is “never” or “years”, change the plan before you spend.
For market size, the estimate suggests customer groups and gives each a TAM, SAM and SOM with a note on what the number covers. Treat them as hypotheses to test, not verified demand. It takes about four minutes to see what the software would cost to build, and it is a planning range, not a fixed quote.
If your business is a startup that has found its market and now needs to grow the team and the product, read our guide to startup scaling.
Usually keeping and selling more to the customers you already have, because you’ve already paid to win them. Then take back hours with automation, so the same team can serve more customers.
Start with one repeat task that eats hours every week. In the US, 19.8% of businesses used AI in early May 2026, and under 20% of firms with four or fewer employees did. In the EU, a lack of expertise was the top reason firms that had considered AI didn’t use it.
Compare the build cost and a year of running costs with what it saves or earns. For automation, multiply hours saved by the cost of an hour. For a new product, divide the build cost by the profit left per customer to see how many customers pay for it.
A goal that is specific, measurable, achievable, relevant and time-bound. For example, “add 40 regular customers by the end of March” instead of “grow”.
Carefully curated content for resourceful Devs, CTOs, and PMs. No spam.