Someone drops a photo into the group with no message underneath it. By the time it reaches you, three people have already asked who is wearing it and nobody has answered. That is the situation this post is for: a photo with no context at all, that also happens to have traveled through at least one round of a messaging app's compression.
Two separate problems live inside that one situation, and they need different answers. Not knowing anything about the photo is not actually a problem for the search. The compression from being forwarded might be.
Does not knowing anything about the photo matter?
No, and this is worth saying plainly because it is the part people worry about most.
Snagfit reads the picture: what garments are in it, their color, their cut, their material. It does not read who sent the photo, who is wearing the outfit, or which app the photo started its life on. A group chat forward with zero context carries exactly the same information into the search as a screenshot taken straight from the original post, because both are, from the search's point of view, just a picture.
So the frustration of "nobody in the chat knows anything about this" describes a real gap, just not one that blocks finding something to buy. The gap only matters if you specifically want to know the person's name or the original post, which is a different goal than finding the clothes.
Why do forwarded photos often look a little worse?
Because most messaging apps compress a photo before sending it, and forwarding adds another round on top of whatever already happened.
When someone sends a photo through iMessage, WhatsApp or Signal, the app typically shrinks the file to save bandwidth. That is a deliberate tradeoff: faster sending and smaller file sizes, at the cost of some image quality. It is usually not noticeable on a single send.
The problem builds up over multiple sends. If person A saves a photo, sends it to person B, who forwards it to a group chat, who then has someone else forward it to a second group, that image has been compressed on top of itself several times. Each round smooths out fine detail a little more, the same way photocopying a photocopy loses sharpness each generation.
What kind of detail actually gets lost?
Texture and fine pattern go first, well before the photo looks obviously bad to the eye.
A knit sweater's weave, a subtle check pattern, a small logo, or a fine pinstripe are all made of detail at a scale that compression targets specifically, because that is exactly the kind of fine variation compression algorithms are built to simplify. A heavily forwarded photo can still look like a recognizable outfit while having lost most of the detail that would make a garment description specific rather than generic.
The visible warning sign is a patterned area turning into a flatter, smoother block of color, or the edges around a person breaking into small square blocks. If you see either of those, some detail is already gone. If the photo still shows clear texture despite having clearly been forwarded a few times, you are probably fine.
Does the app it passed through change anything?
Only in how much compression it applied, not in any way Snagfit treats it differently.
iMessage, WhatsApp and Signal each handle photo compression differently, and the amount also depends on whether the sender picked a high-quality send option where the app offers one. None of that changes how Snagfit reads the final image. The search does not know or care which app carried the photo. It only sees whatever quality survived by the time the picture reaches you.
So there is no app that is "better for searching." There is only more or less compression along the way, and that comes down to how many times the photo was forwarded and whether anyone along the way used a document or full-quality send option.
Is there a way to get a cleaner copy?
Sometimes, and it is worth a quick ask before you search a heavily compressed version.
Most messaging apps have an option to send a photo as a file or document rather than through the normal photo picker, and that option usually skips the compression step entirely. If whoever has the original is still in the chat, asking them to resend it that way, once, gets everyone a clean copy instead of everyone independently forwarding a copy of a copy.
This is worth doing especially when the outfit has a print, a fine knit, or small details like buttons and stitching that you actually care about, because those are exactly what repeated compression erases first.
Does taking a screenshot of the chat help or hurt?
It usually hurts a little, though it feels like the obvious move.
A screenshot captures whatever is currently on your screen, which means it recompresses an image that may already be compressed from the forward. That adds one more generation of quality loss on top of what already happened. Saving the photo directly, through whatever save or download option the chat app offers, keeps one extra round of compression out of the picture entirely.
If a direct save option is not available and a screenshot is the only way to get the image out of the chat, it still works. It is just not the sharpest version available if a cleaner one can be had another way.
What if the photo is too degraded to search well?
Search it anyway, and treat the result with that in mind.
A softened photo does not stop Snagfit from working. It changes what the search has to go on. Broad categories like garment type and rough color usually still come through clearly even in a compressed image. Fine detail like a specific print, a subtle texture, or small trim is what degrades first and can end up too smoothed over to describe accurately.
If the first search comes back vague, for example "patterned top" instead of something more specific, that is often the compression talking rather than a search problem. Asking the original sender whether they still have the uncompressed file, even after everyone else has moved on from the conversation, is worth a try before assuming the photo cannot be searched well.
Why does nobody ever answer "who is this" in the chat?
Because by the time a photo has traveled far enough to reach a chat with no context, the answer has usually already been lost along the way.
A photo tends to move through several hops before it lands in front of you: someone saved it, sent it to a friend, that friend sent it to a partner, the partner dropped it in a group chat for a different reason entirely, maybe just as a joke or an aside. At each hop, the caption or the "this is so-and-so wearing X" note that may have existed at the start gets left behind, because forwarding a photo rarely means forwarding whatever was typed next to it.
By hop four or five, the photo is just an image floating in a chat, detached from any of the context it may once have had. Nobody in the current conversation is being unhelpful. Nobody there ever had the answer to begin with.
Is it worth asking who originally sent it?
Sometimes, mainly to get a better-quality copy rather than to learn who is in the photo.
Tracing a photo back through a chain of forwards to find whoever saved it first is a real amount of social effort, and it does not usually succeed past two or three hops, since most people cannot remember exactly who sent them something days or weeks earlier. If the goal is finding out whose outfit it is, this rarely pays off and is not needed anyway, since the search does not use that information.
Where asking around does pay off is quality. If somebody in the chain still has the photo saved on their own device rather than only the forwarded version they sent along, that copy is likely to have gone through fewer rounds of compression than the one that eventually reached you. One message asking "does anyone still have the original of this" is worth sending before assuming the compressed version in front of you is the best available.
How is this different from a screen recording with no caption?
Related, but not the same problem.
A group chat forward with no context is usually a still photo that has lost quality through compression. A screen recording is a frozen video frame, which comes with its own separate issues: motion blur from freezing mid-scroll, or interface elements like a play button burned into the image. Finding clothes from a screen recording covers that version in detail, and it is common for a photo to have gone through both: screen-recorded first, then forwarded through a few chats afterward.
And if the photo you are dealing with never had any source to begin with, even before it reached the group chat, that is its own separate case covered in finding clothes with no source at all. The two often overlap: a photo nobody in the chat can trace, and low quality on top of it.
What is the actual first move?
Search the photo you have before doing anything else.
Trying to reconstruct who sent it, where it came from, or what the original quality looked like takes real effort and often goes nowhere. Uploading the photo as it currently exists takes a few seconds and tells you immediately whether it is good enough to work with. If the result looks vague or incomplete, that is the signal to go looking for a cleaner copy, not before.