Putting AI to work inside the systems you already run
What to check in your phones, email, records and files before you build, and how to tell when a system should be replaced instead.
Picture a support agent with two screens. On the left, the customer record. On the right, an AI chat window. Every question means copying details across, waiting for an answer and pasting it back. Within a fortnight the right-hand screen is closed for good.
Often the answer isn't a new system. Put AI where the work already happens, so the help shows up in the record, the email or the call your people are already looking at.
Where does AI usually connect?
Most of the systems an organization runs on can share information with other software. In plain terms, that tends to look like this:
- Phones. Calls can go to AI first, with a route to pass the caller to a person.
- Shared inboxes and email. AI can read incoming messages, sort them and draft replies for someone to send.
- Customer and case records. AI can look things up and write a note back, so the history stays in one place.
- Files and policies. AI can answer from your own documents, as long as they are current and stored where it can reach them.
What should you check before anything is built?
Ask your IT lead, or whoever looks after each system, four questions. Can another tool read from it? Can another tool write to it? Who has to approve that access? And is the information inside it accurate enough to act on?
The last question matters most. AI working from old prices or half-filled records will repeat those mistakes at speed.
What if a system won't connect?
Older or heavily customized software sometimes can't share information easily. That needn't end the idea. The work can often start a step earlier, with the email or form that arrives before anything is typed into the old system, and the connection can come later.
Sometimes the system needs attention first. Better to learn that in a short conversation than halfway through a build.
Sometimes the right answer is to replace or add a system. That should be a business decision, made because it helps the work, never because a new tool demands it.
How do you know it fits the way people work?
Sit with the people who do the work for an hour. Watch where they look things up, what they copy between screens and which steps they skip when it's busy. Those moments are where AI help is most welcome, and they rarely show up in a process diagram.
A short list for your next team meeting
- Name the three systems your team opens most often.
- For each, find out who can approve another tool reading from it.
- Pick the screen where people copy and paste the most, and start your thinking there.
For a second pair of eyes, check the fit with your systems and bring your IT lead along. You can also read how a build comes together, or how agents and automation differ.
Have a challenge like this?
Bring it to a Complimentary Business + AI Opportunity Session, 45 minutes with Brian Pinkus, our Founder & CEO.
Pick a time that suits you. It goes straight onto Brian's calendar.