You post the new-in linen dresses on Thursday evening. By Friday lunchtime there are forty Instagram DMs, and most say a version of the same thing. "Do you have this in a 12?" "How long is it on someone 5'4"?" "Can you put one aside till Saturday?" Meanwhile the shop is busy, the phone is ringing behind the till and someone is asking for a different size in the fitting room. A boutique AI sales assistant answers those messages from your own stock list, size notes and returns policy, books styling appointments, and passes anything that needs a person to your team.
This page is for owners and managers of independent clothing boutiques: womenswear, menswear, occasion wear, pre-loved and designer resale, children's clothing. It covers the questions boutiques actually get, what an assistant like Klugon can answer and what it shouldn't, how your returns policy fits the Consumer Rights Act and the Consumer Contracts Regulations, and what it costs compared with an extra person on the shop floor.
Why a boutique AI sales assistant, and why now
Most boutiques sell in three places at once. There's the shop floor. There's Instagram, which for many independents is the real shop window, because that's where new stock gets seen first. And there's a website, sometimes a full online store, sometimes a lookbook with a contact form. Each of those produces questions, and the questions arrive at exactly the moments you can't answer them: when you've just posted a drop, when the shop is full on a Saturday afternoon, and in the evening when customers browse on the sofa.
A missed DM in fashion has a short shelf life. A customer who asks about a size 10 in the green and hears nothing for six hours has often found something similar elsewhere. Unlike a hair salon or a dentist, you aren't selling a slot only you can fill. You're selling a dress that other shops may stock too, and a fast, accurate answer is a large part of why someone buys from you.
An AI assistant doesn't replace your eye for fit or your relationship with regulars. It handles the repetitive layer, the "is it in stock", "what's the length", "do you post" and "can I return it" questions, so that you can spend time on the conversations that need you.
The messages and calls a clothing boutique gets
If you went through a month of DMs, emails, WhatsApps and calls and sorted them, you'd find the same few categories every time.
Stock and size questions
These are the majority. "Do you have the Ottilie dress in a 14?" "Is the navy blazer still in?" "Have you had any more of the striped Breton tops back?" Many come within hours of a post, which means they're about a small number of items. Some are about older stock the customer saw weeks ago.
Fit questions sit next to these: does it run small, is it true to size, what's the inside leg on the wide-leg trousers, how long is the midi on someone 5'3", is the fabric stretchy, is it see-through, does it need a slip. These are the questions your staff answer well from experience, and that an assistant can only answer if you've written the answer down.
Holds, reservations and click-and-collect
"Can you put it aside till Saturday?" is the boutique version of a booking request. Some shops hold for 24 hours, some till close of the next day, some only with payment. If you sell online, there's also click-and-collect: the customer orders on the website and picks up in store, and then asks when it'll be ready, what ID they need and whether someone else can collect.
Returns, exchanges and refunds
"I bought it last week and it doesn't fit, can I swap it?" "I ordered online and it's not for me, how do I send it back?" "The zip broke the first time I wore it." These three sound similar and are legally very different, and staff often answer them inconsistently. More on that in the compliance section below.
Appointments and events
Personal styling sessions, wardrobe edits, bridal or occasion-wear appointments with a dedicated fitting room, alterations drop-offs if you offer them, and private shopping evenings for regulars. These have a time, a length and often a named stylist, which makes them a good fit for an assistant that can book.
Everything else
Opening hours, parking, whether you're open on the bank holiday, whether you do gift cards, whether you post abroad, whether you have a sale on, and messages from brand reps, influencers and people selling things to you. Plus the occasional complaint.
When they arrive
The pattern is predictable even if the volume isn't. Messages spike within a few hours of a new-in post or a restock story. They cluster in the evenings, when people scroll. In the shop itself, Saturday afternoons and weekday lunchtimes are when the phone is hardest to pick up, because that's when the floor is busiest. Across the year, the heavy periods are the start of each season's drops (spring/summer arrives in February and March, autumn/winter in August and September), wedding and occasion season from April to September, the run-up to Christmas party season, Black Friday, and the January sale, when returns and exchange questions pile up alongside sale enquiries.
What a boutique AI sales assistant handles
Klugon's assistant answers on your website chat widget, WhatsApp, Instagram DMs, Facebook Messenger and the phone. On every channel it works from the same information: your product catalogue, your knowledge base (policies, size guides, FAQs) and your booking calendar.
Stock and size answers from your product catalogue
The assistant can search your product catalogue in chat and on the phone. Each product has a name, description, category, SKU, price, currency and a stock count, and you can upload an image. When a customer asks about an item, the assistant looks it up and sees whether it's in stock and how many you have.
Two things are worth knowing before you set this up, because they shape how you do it.
There's no size or colour variant field. A product is one stock count. If you want the assistant to answer "is it in a 12?" from stock data, you list each size you care about as its own product, for example "Ottilie linen midi dress, sage, UK 12", with its own stock number. For a boutique with small size runs (often one or two of each size), that's manageable for new drops and best-sellers. For the rest of the shop, it usually isn't worth it.
Products are entered by hand and don't sync with your till. There's no import and no connection to Shopify POS, Lightspeed, Square or any other EPOS. If a size 12 sells in the shop at 11am, the assistant won't know unless someone changes the number. The products page has quick plus and minus buttons for stock, and low-stock and out-of-stock filters, which makes a two-minute end-of-day check practical. But it's a manual job, and if it slips the assistant will give out-of-date answers.
So most boutiques do a mix:
- New drops and restocks go in the catalogue size by size, because that's where the DMs are, and you update them daily while the drop is selling.
- Everything else lives in the knowledge base as text: what you stock, which brands, the size range each brand comes in, and an instruction that the assistant should offer to check with the team for specific older items.
- Anything sold out gets set to zero rather than deleted, so the assistant says "that's sold out" rather than "I can't find it".
Fit and fabric questions from your size notes
Size guides, measurements and fit notes go in the knowledge base. You can upload PDFs, Word documents, text files or CSVs (up to 20 MB each), paste text straight in, or import a single web page such as your size guide page. Scanned PDFs (images of text) can't be read, so if your brand size charts are scans, retype them or paste the numbers.
The detail that makes this useful is the kind your staff carry in their heads. Write it down product by product or brand by brand:
- "The Ottilie dress runs large. Most customers size down. Length from shoulder is 121 cm, which is mid-calf on someone 5'5"."
- "Brand X jeans are cut small on the waist. Suggest one size up if between sizes."
- "The linen is not lined and is slightly sheer in white and sage. We suggest a nude slip."
- "Fabric composition is on each product description."
On that last point: under the textile labelling rules, the fibre composition is on the garment label, and customers do ask ("is it 100% cotton?"). The assistant can only repeat what you've told it. Put composition in the product description and it'll quote it accurately. Leave it out and you want it to say it'll check, not guess.
Holds and click-and-collect
The assistant can't put a physical item aside, and a conversation doesn't reduce stock. So decide what you want a hold request to become:
- A lead your staff act on. The assistant takes the customer's name, the item, size and colour, and when they'll come in, and records it as a lead. That posts to Slack or Zapier if you've connected them, so whoever is on the floor sees it and pulls the item. Your hold rules (24 hours, till close tomorrow, whatever they are) go in the knowledge base so the assistant states them correctly.
- A collection appointment. Some boutiques create a short "collection" service in the booking calendar, so customers pick a 15-minute window and staff know who's coming when. This works well for occasion wear, where you want a fitting room free.
For website click-and-collect orders placed through your own online store, the assistant can explain the process: how long orders take to be ready, what to bring, whether someone else can collect, and how long you keep uncollected orders. It can't look up an individual order in your Shopify or other store, because there's no store connection. Questions like "is my order ready?" should be handed to the team, who can check.
Styling appointments and events
Bookings use Klugon's built-in calendar. You set your opening hours, each service's length (a 60-minute styling session, a 90-minute bridal appointment), which staff member does it, the minimum notice you need, how far ahead people can book, and closed dates. The assistant offers real free slots and checks for clashes per stylist, so two people can't be booked with the same stylist at once. The customer gets a booking reference in the conversation.
In chat and messaging, the assistant can also reschedule and cancel. On the phone it can book and cancel, but not reschedule; a caller who wants to move an appointment can cancel and rebook, or you can move it in the dashboard.
It doesn't send confirmation emails, texts or reminders. If you want customers reminded about a styling session, that stays a job for you or your existing tools.
Suggesting what goes with it
You can set the assistant to suggest related products, for example a belt with the dress or the matching trousers with the blazer, and choose how pushy it is and how many suggestions it makes. Keep it light. A boutique customer who asks a size question and gets a three-item upsell in reply tends to stop replying.
The everyday questions
Opening hours, directions, parking, whether you do gift cards, whether you post and how much, international shipping, how long delivery takes, whether the sale is on, and what your returns policy is. These are the answers you've typed a hundred times. Write them once in the knowledge base and they're answered the same way on every channel.
What it should hand over to your team
An assistant that tries to handle everything ends up annoying the customers who matter most. Tell it plainly which conversations belong to a person. When it hands over, Klugon marks the conversation as needing a person, shows a "Needs a person" banner in your inbox and posts a notification to Slack or Zapier. It doesn't transfer phone calls and it doesn't stop replying, so the handover message should tell the customer what happens next ("I've passed this to the team, they'll message you here").
Hand these over:
- Faulty goods and complaints. A broken zip, a seam that's come apart, a colour that ran. These involve the customer's statutory rights and your judgement about repair, replacement or refund, plus often photos.
- Refund disputes. Anything where the customer disagrees with what your policy says, or where they mention a chargeback, Trading Standards or a review.
- Order problems. Lost parcels, wrong items sent, "where's my order", anything needing your store's order system.
- Bridal and group orders. Bridesmaid dresses in five sizes, school leavers' prom group bookings, corporate uniform requests. High value, needs a conversation.
- Alterations specifics. If you offer alterations, the assistant can explain the service and book a fitting, but pricing for anything unusual and "can this be taken in two sizes?" need someone who's seen the garment.
- Brand reps, wholesalers and influencer requests. Take a name and contact details, and flag it.
- Anything personal or sensitive. A customer who's had a mastectomy and wants fitting advice, or someone shopping after a bereavement. The assistant should be warm and hand over quickly.
If your assistant can't answer something, it should say so and hand over rather than invent an answer. Klugon's insights page lists "questions without a good answer", which shows you where the knowledge base has gaps.
Three sample conversations
These are samples written for this page. The boutique, the products and the customers are made up, and they don't come from real recordings.
Sample: an Instagram DM after a new-in post
A sample Instagram conversation. The boutique, Wren & Thistle in a made-up Yorkshire market town, is invented.
Customer: hiya is the sage linen dress from last night's post in a 12??
Assistant: Hi! Yes, we have one Ottilie linen midi dress in sage in a UK 12. A note on fit: it runs a little large, and most customers size down, so if you're usually between a 10 and a 12 the 10 may suit you better. We have two of those in sage.
Customer: oh good to know. how long is it? I'm 5'3
Assistant: It measures 121 cm from the shoulder. On someone around 5'5" it sits mid-calf, so on you it would likely fall a little lower, around the ankle bone. It's unlined and slightly sheer in the sage, so we suggest a nude slip underneath.
Customer: can you hold the 10 till saturday
Assistant: We hold items for up to 48 hours, so I can ask the team to hold the size 10 until close on Saturday. Can I take your name, and a phone number in case they need to reach you?
Customer: Jo Barker 07700 900123
Assistant: Thanks, Jo. I've passed your hold request to the team: Ottilie dress, sage, UK 10, held until 5pm Saturday. The team will confirm here once it's set aside. If you change your mind, just let us know.
What happened behind the scenes: the assistant looked up two products in the catalogue (the 12 and the 10), answered fit and length from the knowledge base, and created a lead with the item and the hold date, which posted to the shop's Slack channel. A member of staff pulled the dress and replied to Jo from the Klugon inbox, which can reply on Instagram within Meta's 24-hour window.
Sample: a returns question in website chat
A sample website chat. The boutique is made up.
Customer: I ordered a jumper online last week and it's not really me. Can I send it back?
Assistant: Yes. Because you bought it online, you have 14 days from the day you received it to tell us you want to cancel, and then another 14 days to send it back. You don't need to give a reason. Please return it unworn with the tags attached, using the returns form in the parcel or the returns page on our website. We'll refund the price of the jumper and the standard delivery charge within 14 days of getting it back. Return postage is paid by you unless the item is faulty.
Customer: I've worn it once to try it with jeans at home, is that ok?
Assistant: Trying it on at home the way you would in a shop is fine. If it's been worn out, washed or the tags are removed, we may reduce the refund to reflect that. If you're not sure, the team can look at it when it arrives. Would you like me to ask them to check anything first?
Customer: no it's fine, thanks
This is the answer the boutique wrote in its knowledge base, and it matches the Consumer Contracts Regulations for distance sales. The point is that the assistant gives the online-order answer here, not the in-store exchange policy, because the knowledge base explains the difference.
Sample: a phone call to book a styling session
A sample phone call. The boutique and caller are made up, and the assistant's voice on Klugon is AI-generated.
Assistant: Hello, you're through to Wren & Thistle's AI assistant. How can I help?
Caller: Hi, I've got a wedding in June and I wanted to book one of your styling appointments.
Assistant: Lovely. Our occasion-wear styling sessions are 60 minutes and are free, with a private fitting room. Do you have a day in mind?
Caller: A Saturday would be best.
Assistant: This Saturday is full. Next Saturday, the 14th, I have 10am or 2pm with Hannah, or 11:30 with Priya.
Caller: 10 with Hannah, please.
Assistant: Can I take your name and a phone number?
Caller: Sam Ellis, 07700 900456.
Assistant: Thanks, Sam. You're booked for a 60-minute occasion styling session with Hannah at 10am on Saturday the 14th. Your reference is W-2817. If you know your dress size and any colours you'd like to avoid, it helps Hannah pull options before you arrive. Is there anything else?
Caller: No, that's great.
The booking appears in the shop's booking list and on Hannah's calendar. Nothing else is sent to Sam, so if the shop wants to message her the day before, that's a job for staff.
Setting it up for a boutique
You can set up the website widget yourself in an afternoon. Instagram, WhatsApp and Messenger each need some Meta setup, and the phone needs a number after your trial, so plan those separately.
Setup checklist
- Sign up and create an assistant. Give it your shop name, tone (friendly, a bit chatty, or crisp) and what it's for.
- Write the policies first. Returns and exchanges (separately for in-store and online purchases), holds, click-and-collect, delivery and postage, gift cards, alterations. Paste them into the knowledge base.
- Add size and fit notes. Brand by brand, then product by product for the pieces that get asked about. Include measurements you've taken yourself.
- Add your catalogue. Start with current new-in and best-sellers, one entry per size if you want stock answers. Set the currency to GBP, add photos and write descriptions that include fabric composition and care.
- Set up bookable services. Styling sessions, bridal or occasion appointments, alteration fittings, collection windows. Set lengths, staff, opening hours, minimum notice and closed dates (bank holidays, stock-take days).
- Connect Slack or Zapier. This is how you hear about holds, bookings and handovers. Klugon doesn't email or text the owner, so without one of these you'd be checking the inbox.
- Add the website widget. Paste one script tag before the closing body tag. It works on any site, including Shopify, Squarespace and Wix sites, by pasting the tag; there's no plugin or store sync. See adding a chatbot to a Shopify site.
- Connect Instagram. You need a professional Instagram account linked to a Facebook Page, and a Meta app. Meta requires app review for the messaging permission, and until it's approved only people with a role on your app can message the assistant. Start this early.
- Connect WhatsApp if you use it. You'll need a Meta app, a WhatsApp Business number and an access token. It adds $19 a month.
- Decide about the phone. If your shop phone rings out on busy afternoons, forward it to a Klugon number after your trial (see keeping your number with call forwarding). Write your own greeting; the default doesn't say it's an AI, and many boutiques prefer that it does.
- Print your QR code. Every business gets a public page at klugon.ai/your-name with products, services, reviews and a chat, plus a printable QR code. A small sign by the till or in the window, "message us about sizes and holds", works well.
- Test with your staff. Ask the twenty questions you get most often, in the way customers phrase them, including typos and slang. Fix the answers that are wrong.
The weekly routine that keeps it accurate
The setup matters less than the habit. Each morning, or at close, someone checks stock numbers for the current drop. Each week, someone reads the "questions without a good answer" list and the "most asked questions" list on the insights page and adds what's missing. When you change a policy, change the knowledge base the same day.
What it costs, worked through
Klugon prices three products separately, in US dollars. A boutique typically uses the social agent (Instagram, WhatsApp, Messenger), the website chatbot, or both, and sometimes the voice agent. See plans and pricing for the full tables.
Here's a worked example with stated assumptions. Say your boutique gets around 250 Instagram and WhatsApp conversations a month and 80 website chats, and you don't need the phone.
- Social agent, Starter: $79 a month, 300 conversations. One conversation is one customer thread, not one message.
- WhatsApp number: $19 a month, if you connect WhatsApp.
- Website chatbot, Starter: $49 a month, 100 conversations.
- Total: $147 a month, or less on yearly billing (yearly is 20% off the monthly price).
Set that against people. From April 2026 the National Living Wage for workers aged 21 and over is £12.71 an hour. If you paid someone just to answer messages for 15 hours a week, that's about £190 a week, or roughly £826 a month, before employer's National Insurance, pension and holiday pay. That person would also be useful on the floor, which is the real comparison: the assistant isn't a replacement for a sales assistant, it's a way of not needing one just to keep up with DMs.
And against the sales you'd otherwise lose. Say 10 holds or size questions a week go unanswered for long enough that the customer buys elsewhere, and a quarter of those would have bought a £70 dress. That's about £175 a week in sales you'd have made, or around £750 a month. Your numbers will be different. The insights and analytics pages show conversations, leads, bookings and busiest hours, so after a month you can count rather than guess.
The trial is 7 days. A card is required and is charged when the trial ends unless you cancel. During the trial you can test the assistant in the browser and on your website; phone numbers can be added once the trial has ended.
Returns and consumer law, without the guesswork
This is the part of a boutique's messages where a wrong answer causes real trouble, so it's worth spelling out the rules the assistant should follow. What follows is general information about UK consumer law, not legal advice. Check your policy with a solicitor or your local Trading Standards service if you're unsure.
Bought in the shop and changed their mind
The law doesn't require you to refund or exchange something a customer bought in person and no longer wants, for example because it's the wrong size or colour. The government's guidance on accepting returns and giving refunds says so plainly. Most boutiques offer an exchange or credit note anyway, usually within 14 or 28 days with a receipt and tags attached. That's your policy, and the assistant should describe it as your policy, not as the law.
Bought online, by phone or through a DM
Sales made at a distance, which includes your website and orders taken over Instagram or WhatsApp, fall under the Consumer Contracts Regulations 2013. The customer can cancel within 14 days of receiving the goods without giving a reason, then has another 14 days to send them back. You refund within 14 days of getting the goods back, or of evidence they've been sent. The refund includes standard outbound delivery. You can deduct for loss of value if the customer handled the item more than they would in a shop, which is why "tried on at home" and "worn to a party" get different answers.
One exception matters for some boutiques. The regulations exclude goods made to the customer's specifications or clearly personalised. A dress altered to someone's measurements or a made-to-order piece may fall into this; a standard dress in a standard size doesn't. If you offer made-to-measure, say in the knowledge base exactly which items are non-refundable on change of mind, and have the assistant hand over anything borderline.
Faulty, not as described, or not fit for purpose
Under the Consumer Rights Act 2015, goods must be of satisfactory quality, as described and fit for purpose. A customer who rejects faulty goods within 30 days is entitled to a full refund. After that, they're generally entitled to a repair or replacement first. Signs saying "no refunds on sale items" don't remove these rights for faulty sale goods. The assistant should explain that faulty items are dealt with by the team and hand them over, rather than deciding the outcome.
How to write this into the knowledge base
The single most useful thing you can do is write three short, separate answers: one for in-store change of mind, one for online and DM orders, and one for faulty goods. Label them clearly. Then tell the assistant to ask "did you buy it in the shop or online?" before answering a returns question. In testing, that one instruction removes most wrong answers.
Data protection
Messages contain names, phone numbers, sometimes measurements and occasionally health details. Klugon is operated by a German company under the GDPR, and transfers to its processors are covered by standard contractual clauses, as listed in the privacy policy. As the shop, you're the controller of your customers' data. The ICO's advice for small organisations is the practical starting point. Don't add customers to your newsletter because they asked a size question: the assistant records marketing consent only when a customer gives it explicitly, and you can export or delete records from the dashboard.
How it compares with what boutiques already use
Answering Instagram DMs yourself
This is what most independents do, and it works until it doesn't. The trouble isn't the quality of your answers, which will be better than any assistant's on fit and styling. It's timing. You can't answer forty DMs while serving, and the evening DMs land when you're off. An assistant takes the first reply and the repeat questions, and you pick up the conversations that need you from the same inbox.
Instagram's own auto-replies and saved replies
These are fine for "we're open 10 till 5". They don't know what's in stock, can't book an appointment and can't tell the difference between an online return and an in-store exchange. An assistant reads the question and answers from your data.
Chat tools in your online store
Store chat apps and general website chatbots can work well inside the store they're built for. Klugon's difference is that the same assistant answers on Instagram, WhatsApp, Messenger, your website and the phone, using the same knowledge and bookings, and the conversations land in one inbox. The trade-off is that it doesn't sync with your store's products or orders, so stock is kept up to date by hand. For a side-by-side with a popular website chat tool, see Klugon compared with Tidio.
Hiring a Saturday person
Often the right answer for the floor. Not usually the right answer for the DMs, because the DMs peak in the evening and in the hours after a post, not only on Saturdays.
Getting the most from it in a boutique
A few habits separate boutiques where the assistant is useful from those where it isn't.
- Write fit notes in your own voice. "Runs large, size down" is more useful than a brand's generic chart, and it's the knowledge that makes your shop worth buying from.
- Update stock on drop days. If you only keep one thing current, make it the newest stock, because that's what people ask about.
- Keep sold-out items at zero for a while. Customers ask about pieces they saw weeks ago. "Sold out, but we're expecting the autumn colours in September" is a good answer if you've written it down.
- Use the busiest-hours chart. The analytics page shows when conversations come in. If you see a spike on Sunday evening, that's when new-in posts get the most replies.
- Read the handovers. If the same kind of question gets handed over every week, write an answer for it.
If your shop also does alterations, the page for tailors and alteration shops goes deeper on fittings and turnaround times. Boutiques that also sell gifts, cards and homeware will recognise a lot in the page for gift shops, and if you stock activewear, the sports shops page covers footwear sizing and workshop bookings. For the wider picture across shops of every kind, see AI assistants for retail. The AI sales assistant feature page explains how product search and suggestions work in more detail.
Frequently asked questions
Can the assistant tell customers which sizes are in stock?
Yes, if you list each size as its own product with a stock count. The catalogue has one stock number per product and no separate size field, so "Ottilie dress, sage, UK 12" is one entry. Stock is updated by hand and isn't synced with your till or online store.
Does it connect to Shopify or my EPOS?
No. You can add the chat widget to a Shopify site by pasting one script tag, but there's no product, stock or order sync with Shopify, Lightspeed, Square or any other system. Products are entered and updated in Klugon.
Can it put items on hold for customers?
It can take a hold request with the customer's name, the item and the time, and pass it to your team as a lead, notifying Slack or Zapier. Someone in the shop then pulls the item. A conversation doesn't change stock levels.
Will it give the right answer about returns?
It gives the answer you write. Write separate answers for in-store change of mind, online and DM orders, and faulty goods, and tell it to ask where the item was bought first. Hand faulty-item and dispute conversations to your team.
Can it book styling appointments?
Yes. It books into Klugon's built-in calendar using your opening hours, appointment lengths and which stylist does each one, and gives the customer a booking reference. It doesn't send confirmation emails, texts or reminders.
Does it work on Instagram straight away?
Not immediately. You need a professional Instagram account linked to a Facebook Page and a Meta app, and Meta requires app review for the messaging permission. Until that's approved, only people with a role on your app can message it.
Can I reply to customers myself?
Yes, on Instagram, WhatsApp and Messenger, from the Klugon inbox, within Meta's 24-hour window. You can't reply to website chats or phone calls from the inbox. The assistant keeps replying while you do, so tell the customer you've taken over.
What happens with complaints?
Tell the assistant to hand them over. It marks the conversation as needing a person, shows a banner in your inbox and notifies Slack or Zapier. It can't transfer a phone call, so it tells the customer the team will be in touch.
Can it take payment for an item in a DM?
Don't plan around that. Klugon is built to answer questions, take holds and book appointments. Payment for boutique items is best taken through your own store or in person.
Is there a free trial?
Yes, 7 days. A card is required and is charged when the trial ends unless you cancel. You can test everything in the browser during the trial; a phone number can be added once it ends.
If the DMs are what's falling behind, start with the widget and the knowledge base, then add Instagram once Meta has approved your app. You can read how Klugon's AI receptionist for small business works across every channel, or start a trial and test it on your own size questions.