The temptation with automation is to start with the most exciting thing. An AI assistant that answers everything. A dashboard that predicts the future. In practice, the businesses that get the fastest return start somewhere much more ordinary.
Start with a bottleneck, not a technology
A good first automation sits where work is slow, repetitive and clearly bleeding value. The tell-tale signs are work that is done the same way every time, work that depends on one person remembering to do it, and work where a delay costs you a customer.
For a lot of Kenyan businesses that points to the same place: the gap between an enquiry arriving and someone responding to it properly.
The three questions that pick your first project
- How often does this happen? Daily beats monthly.
- What does it cost when it goes wrong? A lost lead is expensive. A slow internal report rarely is.
- Can it be described as rules? If you can explain what should happen in a sentence, it can usually be automated.
Where most businesses land
Lead handling is usually the first win. Capturing an enquiry, acknowledging it quickly, capturing the details you need, and making sure a follow-up actually happens. It is not glamorous, but it directly affects revenue and it is measurable.
Once that works, the same approach spreads to quotes, support questions, order updates and reporting. Each one removes a little more manual work and a little more risk.
What not to automate first
Do not automate a process you have not fixed. If the underlying process is confused, automating it just produces confusion faster. Map it, simplify it, then automate it.
And do not automate judgement. Pricing exceptions, difficult customers, sensitive conversations. Those stay human. The goal is to remove the repetitive work around them so people have more time to do them well.
