How does a dietitian / nutrition consultant use intusell? A daily operations guide

In the cornerstone article you trained the intusell AI: you uploaded your services and packages, first-consultation preparation instructions, working hours and frequently asked nutrition and price questions to the knowledge base, marked past client consultations, and set the persona and the nutrition-specific boundaries. Training was a one-off setup. Now the real matter begins: how does a dietitian / nutrition consultant use intusell, that is, how does dietitian intusell use work in daily operations? This article shows step by step what the assistant you trained does, and how, throughout a workday in the field — from the "how much is a session?" question coming in via Instagram DM to a first consultation, from a follow-up appointment to client tracking and to human handoff.

This article is the second part of the dietitian chapter of our sector-by-sector "how to train your AI" series. The first part, dietitian AI training, explained the setup; this article explains turning that setup into daily operations.

Quick answer

The flow of a day

The typical day of a dietitian who has set up intusell is very different from before. Previously, when you sat down at your desk in the morning, you would scan dozens of unread Instagram DMs and unanswered WhatsApp messages one by one, sorting out which asked about price, which wanted an appointment, which was a program change. Now, the messages that came in overnight are already answered, requests are pre-classified, some first consultations are already arranged, reminders have gone out. The backbone of a day works like this:

You are no longer occupied with scanning messages and the "is it price or appointment or program" sorting work, but with the work that truly requires expertise — client consultations, drawing up programs, follow-up assessment.

Who is it for?

This usage model is especially meaningful for these nutrition practices:

• Dietitians receiving heavy message volume from Instagram and WhatsApp — those who can't keep up between consultations with dozens of "how much is it, do you do online?" DMs a day.
• Consultants losing the prospect before the first consultation — those with a flow of prospects who ask about price and, because the reply came late, end up writing to another dietitian.
• Practices missing after-hours and evening demand — those who, by the time they see the interest that came in at night the next morning, find the prospect has already enrolled elsewhere.
• Dietitians handling clients, appointments and social media alone or with a small team — those forced to track comments under posts, DMs and the appointment book separately.

If there is a message volume one person can't keep up with, intusell works like a senior client coordinator with clear boundaries who meets that demand — not like a chatbot.

The most important boundary: the AI does not give nutrition or medical advice

For a dietitian, this is the foremost rule. In a concrete nutrition or health situation, intusell does not provide a personalized diet plan, an "eat this / cut that" type of direction, or a medical interpretation. This is both a professional boundary and prevents misdirection from harming the client; the binding assessment is made by the dietitian, and a physician when necessary. Moreover, the AI does not have the client's lab results, chronic illness or medication — so making recommendations would not be appropriate in any case.

In practice, what the AI does and does not do breaks down like this:

When a client asks a concrete question like "I'm diabetic, can I continue with this program?", the AI does not list out a program; it says "Let's assess this together in the consultation, and it would be good to consult your physician too if necessary," and routes to the dietitian. Wherever it doesn't know or expertise is required, it honestly states its boundary and hands off. We explain in detail how you set this boundary during training in the dietitian AI training article.

Incoming request: the client asks, the AI gives the information and routes

The flow starts when a client sends a message. Imagine someone writing via Instagram DM, "Do you provide online nutrition counseling, how many sessions does it take?". The AI reads this message, pre-classifies what the request is (here both a service inquiry and a duration/session question), explains the services and packages from the knowledge base, and routes the prospect to a first consultation or an introductory call.

Two points are critical. First: the AI does not state price and packages off the cuff; it shares only the current service and price information you uploaded to your knowledge base. Second: the AI replies in whatever language the client writes in — a practical convenience for nutrition professionals working with Turkish clients abroad.

This way the prospect who reaches you arrives qualified from the start, and there's no need to interrogate the client over and over. You'll find the details of the appointment and follow-up engine in the appointment and follow-up automation article; how all channels merge into a single inbox we explain in the Instagram and WhatsApp automation article.

Service and price questions: answers from the knowledge base, no inventing

Most of the messages a dietitian receives are not from a new person but the same recurring question: "How much is a session?", "Do you do online consultations?", "How many sessions does it take?", "How does follow-up work?", "Should I come to the first consultation fasting, should I bring lab results?" These questions cycle in the same pattern all day and eat into the gaps between your consultations. intusell takes this load by answering from the knowledge base.

The most critical rule here is this: the AI does not invent service and price information. It gives the answer only from the source you uploaded to your knowledge base:

If information not in the knowledge base is asked about — for example, a campaign you didn't upload or a personalized package — the AI, instead of inventing information, says it doesn't know and offers to connect the client to you. This behavior prevents you from having to give the prospect a wrong price and then say "actually that package is different." A wrong price statement damages both the client relationship and your practice's credibility; that's why the "say when you don't know" behavior is not a constraint but provides protection.

How the knowledge base is fed and stays current

The AI giving correct answers depends entirely on how current your knowledge base is. intusell feeds the knowledge base from several sources and indexes all of them for pgvector-based search (RAG):

• Service and package files: uploaded as PDF, Excel or CSV; service types, session counts, package contents and price ranges come from here.
• Web URL: you provide the address of your clinic/consultancy page, your services page or your FAQ page; the content is fetched and indexed.
• Free text: you write notes like "This month, the first consultation is free in the 3-month online follow-up package" directly into the panel.

For a dietitian, the package arrangement and follow-up structure change from time to time. When you update your prices or package contents, you refresh the knowledge base as well. The AI always speaks according to the most recent source you uploaded — if the source isn't current, the AI isn't either, which is why currency is the backbone of operations.

An additional security layer: the AI is constrained by a sector-specific system instruction; on top of this, a guardrail layer scans responses and flags risky phrases (for example, "guaranteed weight loss," "you'll definitely get results," "you'll slim down with this program"). This layer runs in shadow mode by default — it does not hard-block, it detects and flags; the real behavioral control is in the system instruction.

First consultation and follow-up appointment: slot, working hours and double-booking prevention

When a client requests a first consultation or follow-up appointment, the AI does not offer times off the cuff. It looks at the three things you defined during training: the duration of the appointment type (for example, "First Consultation 45 min", "Follow-up 20 min"), your working hours and whether that slot is booked. If it finds a suitable slot, it creates the first consultation or follow-up appointment. This is the appointment engine's most visible job for a dietitian: "can I come for a first consultation Thursday 16:00?" is resolved with a single message.

The most critical protection here is double-booking prevention: the same slot cannot be double-booked. Before creating an appointment, intusell checks your working hours and currently booked slots; if a conflicting slot is requested, it does not open the appointment. Canceled or no-show (missed) appointments free the slot up again. An automatic reminder (1 day and 2 hours before) goes out for each appointment; this reduces no-shows — a wasted consultation hour touches money directly, and moreover a missed follow-up causes the client to disconnect from the program. The client manages their appointment themselves via the /manage-appointment/{token} link; canceling or rescheduling is done in one click, so the slot opens up for another client. Appointments are copied one-way to Google Calendar (intusell → Calendar); you see them in your own calendar. This module can be enabled on every plan; if you don't need it, you simply don't use it.

Client tracking and conversion: the follow-up engine's most valuable job

For a dietitian, the most frequently lost prospect is the one who asks about price or a program, says "let me think about it" and doesn't come back, and is forgotten because no one followed up. Acquiring a new client is far more expensive than converting a prospect who showed interest; that's why this is the follow-up engine's most valuable job in your practice. intusell tracks prospects who asked about price and went silent or postponed the first consultation with a proactive follow-up flow and sends a timely reminder.

The reminder goes out through the channel the client wrote on; if they can't be reached there, SMS kicks in, and if that fails, email. The message is not a dry "come to us" call; it routes the prospect either to the first-consultation flow or to a short introductory call. For an interested prospect, the AI gathers the necessary information (which package they're looking at, whether they want online or in-person) and hands the prospect to you fully qualified.

A boundary should be drawn clearly: intusell is not a separate product that keeps a managed, personalized "waiting list." What it does is offer an opportunistic re-offer/reminder to suitable WhatsApp prospects in follow-up — for example, a message to someone who asked about price and didn't return, in the tone of "Shall we talk through the online follow-up package you asked about last week?". This way, when you sit down to operations in the morning, it's clear which prospect is warm and which is awaiting a response; you're spared keeping a list by hand and calling one by one. We explain in detail how this engine merges with the appointment flow in the appointment and follow-up automation article.

Health and medication questions: inform, then route to the expert

A client often writes with a health concern: "I have a thyroid condition, is this program right for me?", "I'm pregnant, can I get nutrition counseling?", "I use insulin, is there anything I should watch out for?" At this moment, intusell behaves in a balanced way. It makes no medical interpretation, does not rule "suitable/unsuitable"; it shares the general service information you entered into the knowledge base and which situations you work with, routes the situation to you and, when necessary, says to consult a physician.

In other words, on health matters the AI is an informing and routing layer: it carries the client to the right party without misdirecting them. Determining the program, assessing lab results and any adaptation in case of illness/medication always stay with you. The client feels taken seriously on this sensitive subject; you, meanwhile, take over the situation not from scratch but with the context the AI gathered. As in healthcare, here intusell provides information, not advice, and leaves the binding assessment to the expert.

Client line: the dietitian pipeline

Every prospective client moves through a customer line (pipeline), and the AI carries the prospect to the right stage as the conversation develops. The stages of a typical dietitian line can be thought of like this:

New Interest / Price Question
  → First Consultation Scheduled
  → First Consultation Held
  → Package / Follow-up Consultation
  → Client Enrollment Opened (won)
or → Dropped (lost)

Thanks to this line, you see at a glance which prospect only asked about price, which came to the first consultation, and which is awaiting a package decision. For the post-consultation conversion and regular follow-up tracking side, a separate follow-up flow runs; this way, new interest and existing-client tracking proceed in the same panel but without getting mixed up.

Human handoff: working modes

You determine how autonomously the AI operates. There are three working modes:

In hybrid mode the AI hands off to you a medical/medication question, a personalized negotiation request, or a question outside the knowledge base. At the moment of handoff, the entire conversation history is in front of you; the client does not have to explain the situation from scratch. You can change the mode at any time — for example, ai_only during a busy New Year campaign, human_only for a sensitive clinical case.

An additional note: when a client sends a file such as a receipt image, an old lab result, a health report or an old diet plan, that alone is not a reason to hand off to a human. The AI receives the attachment, understands the context and continues the flow; it hands off only if there is a situation that truly requires a decision (for example, an active health issue, medication, negotiation). The medical interpretation of lab results always stays with you.

Privacy, KVKK and sensitive data

For a dietitian, the incoming information is often sensitive: name, phone, goal, and in some cases health declarations, lab results and medication history. intusell protects this at three points:

1. Explicit consent log: The client's explicit consent is recorded along with its time and type (kvkk_consent_at / kvkk_consent_type).
2. Sensitive field control: Fields such as health declarations, lab results and medication information are flagged with the highest sensitivity; they are not saved automatically and require manual approval.
3. Encrypted storage and masking: Personal data is end-to-end encrypted; PII masking is applied to audio recordings. Data is isolated on a per-tenant basis.

This makes it easier to answer, with evidence, the "what did the client consent to and when?" question in the event of an audit or objection; at the same time, it prevents sensitive health information from being recorded haphazardly.

Reports: what's working, what to fix?

To keep operations visible, you monitor core metrics in the panel. How many messages came in from which channel, how many were answered by the AI, how many were handed off to you, which services and packages are asked about most, how many first-consultation requests turned into enrollments — seeing these serves two purposes:

1. Finding knowledge base gaps: If there's a service or package the AI frequently says "I don't know" about and hands off, that information is missing from the knowledge base; once you add it, the handoff rate drops.
2. Improving response rules: You review the labels and correct mis-classified conversations (for example, those that mistook a first-consultation request for a price question); the AI learns from these corrections. If you like, you compare two different post-consultation follow-up approaches via A/B testing.

In other words, reports are not just a summary but the feedback loop for sharpening the assistant over time.

What it isn't

To place intusell in the right category, let's clarify what it is not:

• It is not a dietitian or health consultant. It does not draw up a diet plan, does not give nutrition recommendations, does not interpret lab results, does not give medical advice; the dietitian and, when necessary, a physician handle these.
• It is not a client/accounting software. It does not enroll the client automatically, does not collect payment; it gathers the enrollment request and carries it to you, and the one who closes the client is you.
• It is not a nutrition tracking dashboard. It does appointments and follow-up tracking with the appointment engine, but does not take on program tracking, macro calculation or clinical record management; it is a communication and coordination layer.
• It is not a product priced specifically for dietitians. The pricing model is based on messages and voice minutes, and does not change by sector. The appointment and follow-up modules can be enabled on every plan.

One more note: Instagram comment automation is rolled out gradually depending on Meta approval; the first step is always WhatsApp and Instagram DM, which are approved. So that you don't start with the wrong expectation, let these boundaries be clear from the outset. For product and price details you can look at the solutions and pricing pages.

Frequently asked questions

Does the AI close a client into a sale on its own?

The AI answers service, package and program questions from the knowledge base and carries a suitable prospect as far as a first consultation or an introductory call. The contract, payment and enrollment stay with you. In a situation outside the knowledge base, requiring negotiation, or involving a medical matter, the AI hands the conversation off to the dietitian. The one who closes the client is always you.

Does the AI give diet or nutrition advice?

No. The AI does not draw up a personalized diet plan, does not give nutrition advice such as "eat this, cut that," and makes no medical interpretation; it refers these questions to the dietitian. It does not have the client's lab results, chronic illness or medication information, so it would not be appropriate for it to make recommendations. The AI here is a layer for informing and routing to the right person; the expert does the counseling.

How does it create the first consultation and follow-up appointment?

The AI looks at the duration of the appointment type, your working hours and whether that slot is booked; if it finds a suitable slot, it creates the first consultation or follow-up appointment. Double-booking the same time is prevented by the system. An automatic reminder (1 day and 2 hours before) goes out for each appointment, which reduces no-shows.

Can the client be connected to the dietitian?

Yes. In hybrid mode the AI runs the normal flow and escalates the conversation to you when needed. In human_only mode all conversations come directly to you; in ai_only mode the AI manages them all. You can change the mode at any time. At the moment of handoff, the entire conversation history is in front of you.

What does the AI do with a health or medication question?

The AI makes no medical interpretation. For a question like "I have a thyroid condition, is this program right for me?" it gives no recommendation; it routes the situation to the dietitian and, when necessary, says to consult a physician. Illness, medication, pregnancy or a special medical condition are not matters for the AI to decide on; it informs and carries the person to the right party.

How are client information and KVKK protected?

Client information is end-to-end encrypted; explicit consent is logged with its time and type. Sensitive fields such as health declarations, lab results and medication are flagged with the highest sensitivity and are not saved automatically. When necessary, the conversation is handed off to you along with its full history. Data is isolated on a per-tenant basis.

How often should I update the knowledge base?

Whenever your price, package contents or working arrangement changes. The AI speaks only according to the most recent source you uploaded; if the source is stale, the answer is stale too. Uploading a PDF/Excel from the panel and adding a free-text note takes minutes and directly lowers the "I don't know" handoffs in your reports.

Next step

If you haven't trained your assistant yet, start first with the dietitian AI training article — the foundation of these operations is built there. If you want to go deeper on the channel side, move to the Instagram and WhatsApp automation article; for the details of the first-consultation and follow-up appointment engine, the appointment and follow-up automation article; you can see the dietitian-specific solution on the dietitian page, and review the whole series from the all posts page.

If you're curious how the same "doesn't give advice, routes to the expert" boundary is built in another health field, clinic AI training offers a good comparison; in tourism, the flow is in the tour agency AI training article.

To see live how it would work in your own practice, Get a demo or write directly to hello@intusell.com. In a 20-minute session we open your inbox together and test the first price question and a first-consultation appointment in the system.