Back to Blog
Guides5 min read

When a Customer Sends a Photo Instead of Typing

A shopper screenshots your product and sends it with no caption. Most bots freeze and ask them to 'type your question.' Here's what changed — the AI now actually reads the photo and answers it.


The Moment a Photo Lands in Your Inbox

Here's a scene that plays out dozens of times a day on busy pages. A customer is scrolling, sees one of your products in an ad or an old post, screenshots it, and sends it to your DMs. No text. Just the image.

What they mean is obvious to any human: "Is this available? How much?" But most auto-reply bots can't see the picture at all. So they fall back on the only thing they know how to say:

"Thanks for your message! Please type your question and we'll help you."

The customer just told you exactly what they want — by sending the photo — and the bot asked them to start over. Some will retype it. Many won't. They move on to the seller who answered.

Why "Please Type Your Question" Loses Sales

Photos aren't an edge case. Across the merchants using Jawab24, images are the single most common type of rich message customers send — more than voice notes, stickers, or shared posts. People shop with their thumbs. Screenshotting is faster than describing.

Every "please type your question" reply is a tiny tax on the customer's patience at the exact moment they're most ready to buy. On a page getting a few hundred messages a day, those small frictions add up to real, quiet losses — sales that never show up as a complaint, because the customer just stops replying.

The fix isn't a better nudge. It's for the bot to actually read the picture.

What Jawab24 Does With a Customer's Photo

When a customer sends an image in a DM, Jawab24's AI now looks at it, understands what's in it, and folds that into the same reply it would give any text question. A few of the common cases:

A product screenshot → a real answer

A customer sends a screenshot of your product — no words. The AI reads what's in the image (the product, any text or price shown in it) and answers from your Business Info: whether it's available, the price, delivery. No "type your question" — a direct reply, the way a good salesperson would glance at the screenshot and say "Yes, that one's 160, free delivery."

A photo, then a follow-up question

Customers rarely send everything at once. They send a photo, then a beat later type "how much?" The AI keeps the picture in context, so the follow-up resolves to the product they pictured — not a confused "which item do you mean?"

A receipt or an off-topic image → a human, not a guess

Not every photo is a product. Sometimes it's a payment receipt, an ID, or something completely unrelated. Here the smart move is not to invent an answer. Jawab24 reads the image factually, recognizes it doesn't match a sales question, and routes it to you — a real person — instead of confidently guessing. Reading the photo also means the bot knows when not to pitch.

Which Channels This Works On

Photo understanding works everywhere your customers already message you:

  • Facebook Messenger DMs
  • Instagram DMs
  • WhatsApp DMs

If auto-reply is handling your direct messages on a channel, it now handles the photos in those messages too.

What It Doesn't Do (Yet) — On Purpose

We'd rather tell you the edges than let you discover them:

  • Shared posts: when a customer uses the Share button to send one of your posts, the AI reads the post's caption, not the image inside the shared post. (A customer who screenshots the same post and sends it as a plain photo is covered — that's a normal image.)
  • One clear photo works best. If several images arrive in a single message, the AI focuses on the first one.

These are deliberate limits, not bugs — and honest ones. We'd rather ship the case that helps most merchants today than over-promise.

Your Customers' Photos Stay Private

This matters, so it's worth being plain about: Jawab24 does not store the image file. The AI reads the photo, turns it into a short text description, and keeps only that description in the conversation. The picture itself isn't warehoused. Your customers' photos aren't sitting in a database somewhere.

Nothing to Set Up

There's no new switch to flip. Photo understanding is on by default wherever your auto-reply is already active — exactly like voice-note transcription. If the bot is answering your DMs, it's already reading the photos in them. It answers from the same Business Info you've already written, so the more complete that is, the better the answers.

The Bigger Picture

The whole point of an auto-reply is to answer the customer the way they actually reached out — not to force them into the one format the software happens to understand. Voice notes taught us that. Photos are the next, bigger piece: they're how people shop now.

A customer sending a screenshot at 11pm shouldn't hit a wall. They should get an answer. Now they do.

Keep Reading

Ready to Automate Your Replies?

Start your free trial — no credit card required.

Try Jawab24 Free

© 2026 Jawab24