> ## Content Index > Fetch the complete content index at: https://intelligence.artificialrealitymedia.com/llms.txt > Use this file to discover other available public pages before exploring further. # Automated check-ins are the part that fails - URL: https://intelligence.artificialrealitymedia.com/why-automated-client-check-ins-get-ignored/ - Published: 2026-09-02T12:09:00.000Z - Updated: 2026-09-02T12:09:00.000Z - Description: Dr. Nikki Siso automates the prompting and the monitoring, and keeps a human in the chain. What that means for anyone building accountability software. - Author: Galina Fendikevich - Tags: Health & wellness, #For AI Buyers, #For AI Companies, #What's not working, #Import 2026-09-03 05:20 Fully automated check-ins get ignored because there is nobody at the other end to be accountable to. Dr. Nikki Siso, a certified holistic health practitioner and founder of Conscious Kitchen, built a fifteen-skill AI assessment on [Claude](https://claude.ai/?ref=intelligence.artificialrealitymedia.com) and still keeps a human visibly in the loop. Her reason, given to Artificial Reality, the AI buyer-evidence podcast hosted by Galina Fendikevich, is the most useful thing said about this category all season. > "When it's just a hundred percent AI, then you're just like: I can ignore it today." ## Why do automated check-ins stop working? Because the active ingredient is being witnessed, not being asked. Artificial Reality host Galina Fendikevich framed the problem from the client side: AI can send check-in notifications asking whether you ate well or changed a product, but they are easy to ignore, or to see without registering. Siso's diagnosis went further. > "There has to be a human at some point in the chain. So even if AI did the automatic prompting of, hey, what'd you eat today — your reply, if you knew, was gonna get seen by a human, then it challenges your integrity to not reply. That's what's missing." Read that carefully, because it is a specification. Not persistence, not personalization, not better notification timing. The mechanism is the client's knowledge that a specific person whose opinion they care about will read the answer. ## What does she automate instead? The monitoring and the delivery. Clients on her detox program report a daily score, and a low energy score routes to a specific response rather than to a log. > "I've set up AI when someone's going through our detox program that if they check a score of like feeling really tired today, I know: okay, they're going through their detox symptoms have started, and I can check in on them and be like: all right, today rest." > "All of that can be automated, that I get the prompt when they need me." Delivery is automated too, without losing the person: > "I've set up the daily text messages, and I make the text messages in my voice. So like: good morning sunshine. And I can even send a voice memo that's pre-recorded so they actually hear me. So there are ways to make it still me." And the cost of doing it that way, which is where the product opportunity sits: > "All of that takes hours still to set up. It takes the diligence. I have to go through every single text message, make sure it's all there, it's all like me." ## What would a product in this category have to do? Six requirements come straight from the operator. Three more are additions from Artificial Reality, labelled as such. **Stated by Dr. Nikki Siso:** - Keep a human visibly in the chain, so the client knows a person will see their reply. - Automate the prompting, so the practitioner does not have to remember to ask. - Flag when a client is struggling, so the practitioner is alerted rather than monitoring. - Route by response content: a low energy score during a detox means something specific and should trigger specific guidance. - Carry the practitioner's voice, including pre-recorded audio where possible. - Cut the hours currently spent hand-writing and reviewing every message for tone. **Added by Artificial Reality:** - Escalate silence, not only bad answers. A client who stops replying is the one most in need of contact. - Distinguish a hard week from disengagement, because the correct response is the opposite in each case. - Never let a client believe a human read something a human did not read. ## Where is the line a product should not cross? That last one is an ethical line, and the mechanism creates a real temptation to cross it. If the accountability comes from the client's belief that a person reads the reply, then a product that implies human review it does not deliver would work, briefly, and then destroy the relationship it was borrowing against. In a practice built on people disclosing personal health information, that is not a feature trade-off. ## What should a builder take from this? The market has been pointed the wrong way. Products in this category sell the removal of the coach from the loop, and the coach is the loop. The design principle that fits the evidence runs in the other direction: automate everything except the presence, so the practitioner has more of themselves to spend on it. Siso's own setup already does this. The system asks, watches and flags; she arrives when it matters; her clients still get her, and she does not have to remember. That is a smaller product than an AI accountability coach and a much easier one to sell, because it does not have to argue with the buyer's experience of their own clients. If you are building here, the question to answer on your pricing page is not how much of the coach you replace. It is how many of the hours Siso spends checking tone you give back. ## Where this comes from S1E7: How a non-tech founder built a 15-agent AI business on Claude — the full interview with Dr. Nikki Siso. Listen or watch: [YouTube](https://www.youtube.com/watch?v=R%5FzWAgAEIMw&ref=intelligence.artificialrealitymedia.com), [Spotify](https://open.spotify.com/episode/08s9u7LSBTtGGx94b4z2MZ?ref=intelligence.artificialrealitymedia.com) or [Apple Podcasts](https://podcasts.apple.com/us/podcast/how-a-non-tech-founder-built-a-15-agent-ai/id6780699025?i=1000777283240&ref=intelligence.artificialrealitymedia.com).