Can an AI assistant answer shade and colour-match questions in Instagram DMs?

Yes — when it's grounded in your real catalogue. Sauzy answers shade, tone, and finish questions from your live Shopify product data, so it references your actual shade range rather than a generic guess.

How does it handle questions beyond product facts, like skin concerns?

Sauzy answers from your product catalogue — shades, ingredients, price, stock. When a message falls outside that (a health or skin-condition question, a complaint, or anything it can't answer confidently), it hands the conversation to you rather than guessing. Health advice is a category to keep in your own hands, and the assistant is built to stay inside what your product data supports.

How does it avoid recommending a sold-out or reformulated product?

Every answer traces back to your live Shopify product and inventory data at the moment of reply, not a memory of the range — so a discontinued shade or sold-out hero product is never accidentally promised.

Guide

Instagram DMs for Skincare & Cosmetics Brands: Answering Shade, Ingredient, and Routine Questions at Scale

20 September 2026

Beauty is one of the most DM-driven categories on Instagram, and also one of the least forgiving. A clothing question that gets a slightly-off answer is an annoyance; a skincare question answered wrong — the wrong shade, a wrong ingredient claim, a "yes this is fine for sensitive skin" that isn't — turns into a return, a bad review, or worse. The questions come in fast, they're specific, and getting them right matters more than usual.

The questions a beauty buyer actually asks

Before you automate anything, it's worth being honest about what people are really asking in the DMs:

Shade and colour match. "Which shade for medium/olive skin?" "Is this more warm or cool toned?" A generic reply that doesn't reference your actual shade range reads as useless immediately.
Skin type and concern fit. "Is this okay for oily/acne-prone/sensitive skin?" "Will this break me out?" These are the highest-stakes questions — and the ones where an over-eager "yes!" causes the most damage.
Ingredients. "Does this have fragrance / retinol / alcohol?" "Is it fragrance-free?" These have real answers that live on the product, not in a canned response.
Routine and pairing. "Can I use this with my retinol?" "What order do I apply these in?" This is where an assistant that knows your catalogue can genuinely upsell — the cleanser, the moisturiser, the SPF that go together.
Restocks. "When is [sold-out hero product] back?"

Where accuracy matters more than speed — and where it's the reverse

Most beauty DMs split cleanly into two piles. The first is factual and catalogue-grounded: shade availability, ingredient lists, price, what pairs with what. These should be answered instantly and from your real product data, because the answer either exists in your catalogue or it doesn't — there's no judgement involved.

The second pile is different: anything that edges toward a skin-condition or medical claim. "Will this clear my eczema?" is not a product question, and it's not one you want answered from a product catalogue at all. This is a category to keep in your own hands as a matter of policy — a good automated assistant should stay inside what your product data can support and leave health advice to you.

Keep ingredient and shade answers tied to the real product

The failure mode specific to beauty is confidently stating something about a formulation that isn't true — an ingredient that was reformulated out, a shade that's discontinued, a "fragrance-free" that isn't. If you're answering from memory across a growing range, that drift is almost guaranteed over time. Every factual answer should trace back to the actual product record, not a mental model of what the range used to be.

Turn a matched shade into a sale without the hand-off

Once someone knows their shade — or which serum fits their concern — the moment to close is right then. Making them leave the DM to "go find it on the site" is where the momentum dies. A direct link to the exact product, and ideally the exact shade variant, keeps the conversation moving straight to checkout. (We wrote about that hand-off problem in more detail in Checkout inside Instagram DMs.)

Where this gets hard by hand

One founder answering shade questions from their phone can be excellent at it. The problem is consistency across volume: the same accuracy at 2am, across every DM, comment, and Story reply, on a range that keeps changing. That's the gap Sauzy is built for — it answers from your live Shopify catalogue (shades, ingredients, price, and real stock, not a memory), links straight to the right variant when someone's ready to buy, and hands the conversation to you when it can't answer confidently or a message falls outside your product scope — a request for a human, a complaint, or a question your catalogue can't speak to.

If you sell on Shopify, the Shopify integration page covers exactly what gets synced and what stays under your control.