This blog post is the result of some work that I did for a client, betula.ai.
Five Business Lessons from Betula
When I look at an AI tool as a business consultant, my first question is not, “What can it do?” It is, “What problem is taking more of the owner’s time and attention than it should?”
That is what makes Dr. Richard Block’s experience with Betula worth examining. According to the case study, Richard was receiving as many as 20 to 40 spam calls a day while running Business Growth Accelerator and caring for his wife. Betula’s AI voice assistant, Lucy, helped him handle incoming calls and routine follow-up, saving him an estimated 30 to 60 minutes a day.
Those are Richard’s reported results, not a promise of what every business will achieve. Still, his experience offers five useful lessons for solo entrepreneurs and small-business owners considering an AI receptionist.
1. Start with the interruption, not the technology
Richard did not need AI simply because AI was available. He needed a way to stay reachable without stopping what he was doing every time the phone rang.
His business depends on conversations with strategic partners and prospective clients. Yet unwanted calls were competing for the same attention as workshop development, project work, and caregiving.
From a consultant’s perspective, this is where the diagnosis should begin. Before comparing software features, identify the recurring disruption, how often it happens, and what it prevents you from doing.
The practical question is not, “Should I get an AI receptionist?” It is, “Is managing incoming calls a significant enough problem to justify changing how I handle them?”
2. Protect relationships, not just your calendar
Blocking interruptions is only half the job. The other half is making sure the right people can still reach you.
Richard wanted Lucy to recognize important contacts, explain his business, and take useful messages. When a close friend called without leaving a message, he adjusted Lucy’s instructions so that future calls from that person would receive the attention he wanted.
That detail matters. A call-handling system that saves time but makes valuable contacts feel ignored would be solving one problem by creating another.
My recommendation would be to define your call-handling rules before implementation. Decide who needs priority attention, what information a new inquiry should include, and when a caller should be directed to you rather than left in an automated process.
The goal is selective accessibility, not isolation.
3. Treat onboarding as business design
One of the more revealing parts of Richard’s experience was the access he had to Betula’s team during setup. He could explain his business model and provide his landing page as context.
That helped Lucy understand that Business Growth Accelerator identifies organizational pain points, develops targeted workshops, and connects businesses with relevant experts. The setup then evolved as Richard encountered new situations.
I see this as more than technical configuration. It is an opportunity to clarify how your business should respond when someone makes contact.
A useful onboarding brief would describe your services, ideal inquiries, priority contacts, message requirements, and escalation rules. I would also specify what the assistant must not promise or decide without your approval.
Richard suggested a guided questionnaire to simplify onboarding. From a consulting standpoint, that is a sensible way to make the process more consistent without assuming every business works the same way.
4. Measure completed work, not impressive features
Richard’s estimated daily time savings provide a useful starting point for evaluating value. But one example in the case study shows why minutes alone do not tell the whole story.
After Captain Allen of the Austell Police Department called about a public-safety luncheon, Lucy helped connect Richard with him by phone. Rather than merely recording a message, the assistant helped move the follow-up forward.
For an owner, that distinction is important. A tool can appear helpful while still leaving you with all the work of sorting messages, making decisions, and chasing the next step.
I would evaluate an AI receptionist against practical outcomes: fewer unnecessary interruptions, accurate messages, timely handling of important inquiries, and less unfinished follow-up. Then I would compare those benefits with the subscription cost and the time required to review and manage the system.
Time saved has value, but it is not automatically additional revenue. The return depends partly on what you can do with the capacity you recover.
5. Run a real-world pilot before expanding
Richard’s advice was straightforward: try the service for a month and see whether it fits your business. I would take that advice and give the trial a clear scorecard.
Start with a limited role, such as answering calls you cannot take personally. Record your current call volume and interruptions, then review the assistant’s performance against that baseline.
Richard also identified areas he wanted improved, including notification speed, visibility into dropped calls, contact importing, and calendar support. Those requests are a reminder to distinguish demonstrated capabilities from desired improvements.
During a pilot, test the difficult cases as well as the easy ones. Does the assistant capture unfamiliar names correctly? Can important callers get appropriate attention? Does follow-up happen reliably?
My takeaway is simple: adopt the tool because it improves a defined part of your business, not because a demonstration looks impressive. Before buying, write down the three outcomes that would make a trial worthwhile.
Frequently Asked Questions
Is an AI receptionist a good fit for every small business?
Not necessarily. Based on this case, I would first consider it for an owner who receives frequent interruptions, lacks dedicated call coverage, and needs to remain accessible. A business with few incoming calls or unusually sensitive inquiries may need a different approach.
Can I expect to save an hour a day?
Richard estimates that Lucy saves him 30 to 60 minutes daily, but that is his experience rather than a guaranteed outcome. Your results will depend on call volume, caller needs, setup quality, and how much oversight the assistant requires.
What should I prepare before trying an AI receptionist?
Prepare a plain-language description of your business, common caller questions, priority contacts, and clear instructions for messages and escalation. Also decide what information the assistant may collect, what it must leave to you, and how you will judge whether the trial has succeeded.
This post was created using Wispr Flow and Perplexity.