Playbooks · 5 min read

Fashion on WhatsApp: A Size Chart Is Not a Fit Promise

The PostEngage teamEngineering and support ·

Sell clothes and your inbox has one shape. Before the order it is "will this fit me". After the order it is "it did not fit me". Everything else — fabric, wash care, delivery, restock — sits around those two, and both of them are harder than they look.

A size chart is data. A fit is a judgement.

Here is the distinction that decides how you build.

The size chart is a fact about the garment. Chest 38, length 27, shoulder 16, and whether the measurement is of the body or of the flat garment, which is the detail brands forget to state and customers guess wrong. All of that is fixed, published, and perfectly safe to send automatically, in full, every time somebody asks.

Fit is a fact about a person. Whether a size M works for a specific customer depends on their measurements, their build, how they like things to sit, and what the fabric does — a rigid cotton and a stretch blend in the same nominal size are two different garments.

An automation has none of that. What it has is a fluent way of writing sentences, which means asked "will M fit me?" it will produce something confident and specific, and it will sometimes be wrong. The cost of being wrong is a return, a refund, and a customer who feels misled rather than unlucky.

Send the numbers and let the customer decide. Never let software decide on their body.

So the rule is: any message asking for a recommendation rather than a measurement stops the automation and gets a person. will it fit, which size, size suggest, M lena chahiye ya L, mera size, loose hoga, tight — harvest the real strings from a week of your own messages, because every brand's customers have their own vocabulary for this.

The photograph makes it worse

The way this question actually arrives, more often than not, is as a picture. A customer sends a full-length photo of themselves, or of a dress they already own, and asks which size to take.

Nothing in this class of tool reads images. To an automation that message is empty. It will either sit unanswered or fall through to a catch-all, and a generic "here is our size chart!" under a photo somebody was slightly embarrassed to send is a bad outcome that costs a sale.

Returns, exchanges, and the sentence that causes complaints

The post-purchase half is dominated by size returns, and this is where wording does the damage.

"Easy returns" and "hassle-free exchange" are marketing phrases that customers hear as policy. Then they discover the tags had to be attached, or that the pickup takes a week, or that an exchange means a fresh order once the first one is back. The gap between the phrase and the policy is where the angry message comes from.

The automated version should be boring and complete: the window in days, whether tags and packaging are required, who bears the pickup cost, what happens with a sale item, and how long a refund takes to appear once processed. If any of that is unpleasant, saying it early is cheaper than saying it late.

Safe to automate

The size chart, in full, with a note on whether measurements are body or garment. Fabric and composition. Wash care. The returns and exchange policy as written. Whether an item is currently in stock. Delivery areas.

Route to a person

Which size to take. Any photograph. A damaged or wrong item. A refund that has not appeared. Anything involving a bulk or wholesale enquiry, where prices are negotiated.
Messages held for a person to answer, each labelled with the reason the automation declined.
For a clothing brand the held queue is mostly two things: size advice and photographs. Both are exactly where a human earns the sale.

Drops, restocks and the message that ages badly

Fashion runs on dated content in a way that most retail does not — a drop at 6pm, a sale that ends Sunday, a restock in one colour.

Two things follow. First, any template containing a date, a price or a "today" has a shelf life, and the brand that does not switch it off is the brand telling customers in September about an August sale. Put dated messages in temporary automations with an end date, and keep the evergreen set free of numbers.

Second, a restock or drop announcement is a marketing message on Meta's platform, whatever it feels like to you, and marketing messages need marketing consent and an approved template. Someone asking about a returns policy has not opted into your drop calendar. The exact categories and what each requires are Meta's to define and they change; read their documentation when you set up rather than trusting a summary.

What we actually run today

PostEngage answers Instagram comments and DMs through the official Graph API. That is the entire product — there is no WhatsApp in it, on any plan or behind any flag. This post is about the channel because clothing brands search for exactly this and deserve a straight answer rather than a pitch.

For the channel we do run, the mechanics are fixed: a comment stays answerable for seven days, a DM thread for twenty-four hours, and only the customer's own next message restarts that clock. Templated replies are free and unlimited — a credit is spent only when the AI writes new text — so the size chart going out four hundred times a month costs nothing. Free tier is 100 credits with no card, and packs start at ₹499.

Continue with the Instagram version for fashion, the D2C inbox for the pre-purchase and post-purchase split, and boutiques if you sell from a single store rather than a website.

One email when we publish.

No drip sequence, no “quick question” follow-up. Unsubscribe is one click and we honour it immediately.