Your last five posts probably had a killer hook, on-brand copy, and a scroll-stopping photo. They almost certainly also had zero alt text, a hashtag that reads like alphabet soup to a screen reader, and a caption that opens with six emoji in a row. Nobody flagged it in review, because "accessibility" was never on the content calendar next to "caption" and "hashtags" in the first place.
Here's the part that should make this an easy fix instead of a big project: you don't need a new tool, a new budget line, or your dev team to close most of this gap. Every major platform already ships the fields and editors you need — an alt text box, a caption editor, a hashtag field that doesn't care how you capitalize it. Most teams just never turn them on. Here's the checklist to actually run before your next post goes out.
The Stat: WebAIM's alt text guidance is blunt about the standard: describe the meaningful content of an image, not the filename and not "image of ___." Most social composers will silently fall back to whatever's in the file name if you skip the alt text field entirely — the exact failure that guidance exists to prevent. (Source: WebAIM)
Why This Belongs on the Content Calendar, Not the Dev Backlog
It's easy to assume accessibility is a website problem — something for the dev team to handle with contrast ratios and ARIA labels. But a huge share of a brand's daily content output never touches the website at all. It lives entirely inside Instagram, LinkedIn, X, and TikTok, and every one of those posts is either usable by a screen reader user or it isn't, based entirely on choices made in the composer window, not in code.
Screen readers rely on whatever text is actually present in a post — the alt text field, the caption, the on-screen captions in a video — to build a picture of what's being shown. Skip those fields, and a screen reader user gets a description-free photo, a silent video, and a hashtag read aloud as one long, meaningless string of letters. None of that requires a developer to fix. It requires whoever's scheduling the post to spend about ninety extra seconds per piece of content.
The Six-Point Checklist, Explained
1. Write Real Alt Text — Not the Filename
Every major platform now has a dedicated alt text field: Instagram and Facebook tuck it under "Advanced Settings" in the composer, X has an "Add description" prompt directly under the image, and LinkedIn offers it in the image editor before you post. Use it, and describe what's actually in the frame — the product, the setting, the people, the relevant detail — the same way you'd describe it out loud to someone on the phone who can't see it.
WebAIM's guide to writing alt text is the clearest reference for getting this right: keep it concise, skip "image of" (a screen reader already announces it's an image), and focus on the content that matters in context, not a generic label. If you leave the field blank, some platforms and third-party schedulers will auto-fill it with the raw filename — which is how a screen reader ends up announcing "IMG-4821" in the middle of someone's feed.
2. Caption Every Video, Even the Short Ones
A 12-second product demo needs captions just as much as a 10-minute tutorial. Auto-caption tools are now built directly into Reels, TikTok, YouTube Shorts, and LinkedIn's native video uploader, which means there's rarely a real excuse to skip this step — the barrier used to be cost and tooling, and that barrier is largely gone.
W3C WAI's guide to captions and media accessibility lays out why this matters beyond screen reader users specifically: captions also serve Deaf and hard-of-hearing viewers, plus the large share of viewers who watch social video with the sound off for entirely unrelated reasons — a commute, an open office, a sleeping kid in the next room.
3. Use CamelCase in Multi-Word Hashtags
This is the single cheapest fix on this entire list, and it costs zero extra characters. Capitalize the first letter of each word inside a hashtag — #AccessibleDesign instead of #accessibledesign — and screen readers' text-to-speech engines will read it as two clean, separate words. Skip the capitalization, and that same hashtag gets parsed as one unbroken string, which the text-to-speech engine either mangles into a nonsense sound-out or skips entirely.
This isn't a stylistic nitpick; it's a documented quirk of how screen reader software processes text. Capital letters function as word-boundary markers in a way lowercase runs of text simply don't. #NewProductLaunch reads cleanly. #newproductlaunch does not. Make it a hard rule for whoever's writing hashtags on your team, not a "when I remember" habit.
4. Don't Open a Caption With a String of Emoji
A caption that starts with six celebration emoji in a row forces a screen reader user to sit through "party popper, party popper, party popper, party popper, party popper, party popper" before they hear a single word of actual content. Every emoji has a spoken name, and screen readers read every one of them, individually, every time.
That doesn't mean cut emoji entirely — used sparingly and placed at natural pauses or the end of a thought, they're harmless and often add genuine tone. The fix is just sequencing: lead with your actual words, and let emoji punctuate rather than open the caption.
5. Manually Fix Auto-Captions When They're Wrong
Auto-captions are a real improvement over nothing, but they're not accurate enough to trust blind, especially with brand names, product names, and industry jargon. "WCAG" reliably becomes "wick-ag" or "double-u cag" in auto-generated captions unless someone edits it. The fix here is a five-minute manual pass before you publish, not a full re-transcription — just scan for the words a generic speech-to-text model wouldn't know: your brand name, your product names, and anything technical.
6. Describe the Image in the Caption Too
Alt text is essential, but it isn't the only place a description should live, and it isn't always where you'd expect. It gets stripped in reposts, doesn't carry over into most link previews, and isn't visible in every third-party embed or aggregator that pulls your content. Putting a short, natural description of the image in the caption itself — not a duplicate of the alt text, just a plain-language mention of what's shown — adds redundancy that protects the message even when the alt text doesn't make the trip.
This overlaps with a broader idea in accessibility: giving every piece of content a clear, descriptive accessible name wherever it appears, not just in one hidden attribute. MDN's accessible naming guidance covers this concept in more technical depth for web elements, but the underlying principle is identical whether you're labeling a button or a Reel: don't make the description depend on one single field surviving the trip to the reader.
Where Alt Text and Captions Actually Live, Platform by Platform
| Platform | Alt text location | Native caption editor |
|---|---|---|
| Advanced Settings → Write Alt Text | Yes, auto + manual edit on Reels | |
| Edit photo → Alt Text | Yes, auto + manual edit on video | |
| X (Twitter) | "Add description" under image | No — captions require pre-burned or uploaded SRT |
| Image editor → Add alt text | Yes, auto + manual edit on native video | |
| TikTok | Not available for images (video-first platform) | Yes, auto-captions with manual edit |
| YouTube Shorts | N/A | Yes, auto-captions with manual edit |
None of this requires new software, a bigger budget, or waiting on engineering — it requires making these six checks a standing part of whoever schedules your posts' pre-publish routine, the same way spell-check already is. If your website hasn't gotten the same level of attention as your feed, that's worth a look too — you can get a free accessibility scan of your site and see exactly where it stands.