<- Blog
Founder marketingJuly 27, 20268 min read

How to Design a Social Link Preview Card People Want to Click

A practical founder guide to making social link previews instantly understandable, relevant, trustworthy, and worth clicking.

DistributionOS social preview card reading Stop shipping apps nobody sees.
A purpose-built preview gives the page one clear promise before the click.

Every time somebody shares your website, your site gets a small pitch inside somebody else's feed.

It appears on X. LinkedIn. Facebook. Slack. Bluesky. Messages. Group chats. Community servers.

Most founders spend days polishing the homepage, then let the preview use whatever screenshot, logo, or random image the website happened to find.

DistributionOS made the same mistake.

When the DistributionOS homepage was pasted into X, the preview showed an oversized product screenshot with text cut off at the edges.

It technically worked. It just gave a stranger no reason to care.

The original DistributionOS social preview using a cropped product screenshot

That is the missed lesson: the link-preview card is not just website setup. It is a marketing asset.

The card is the pitch before the page

Someone has to choose your link before your homepage gets a chance to explain anything.

The card often has only a few seconds to help them:

  • understand what the page is about
  • recognize that it is relevant to them
  • feel enough interest or trust to stop scrolling
  • decide that opening the link is worth it

The goal is not to summarize the entire website.

The goal is to make the right person think, "This is for me. I understand what I will get. I want to see more."

That makes the card much closer to a small ad or landing page than a decorative image.

The four jobs of a useful card

A good card does four things.

1. Get attention with one clear problem or outcome

Lead with the one thought you want someone to understand immediately.

You can name a painful problem:

Stop shipping apps nobody sees.

Or promise a useful outcome:

Know what shipped. Learn what worked.

Do not stack several messages together. One sharp promise is easier to understand than a headline, subheadline, paragraph, and feature list squeezed into the same image.

2. Help the right person recognize themselves

Attention is not enough. The card needs to feel relevant.

Use a visual that supports the promise:

  • a clean crop of the real product
  • a simple workflow
  • a before-and-after
  • the result the customer wants
  • a familiar situation that reflects the problem

The visual is not there to decorate the headline. It should help a founder see how the message connects to their own situation.

3. Make the source clear and trustworthy

Include enough of the brand or product to show that the link leads somewhere real.

A small logo or wordmark is usually enough. A real product element can add proof.

The logo should answer "Who made this?" It should not become the main message.

4. Give them a reason to take the next step

The card should make the click feel useful.

That might be a simple action:

  • See how it works
  • Read the guide
  • Try it free
  • View the launch
  • Get started

You do not always need a button. The headline and visual can create the desire to click on their own.

The important question is: does the card make the next step obvious and worthwhile?

How to build the message

Start with the page, not the template.

Step 1: Choose the person

Who should stop for this link?

Do not answer "everyone." Name the founder, buyer, customer, or user who should recognize themselves.

Step 2: Choose one promise

What is the single most useful reason for that person to open this page?

Write several versions. Keep the one that is clear without needing an explanation.

Step 3: Choose one supporting visual

Show the product, process, problem, or result that makes the promise believable.

If the screenshot only makes sense when somebody reads tiny labels, choose a different crop or a simpler visual.

Step 4: Add restrained branding and one action

Add the logo, product name, and next step without letting them crowd the message.

Step 5: Test it at feed size

Shrink the card.

Can someone understand it in two seconds? Does it still work without reading the post that surrounds it?

If not, remove something.

Do not turn it into a tiny landing page

The easiest mistake is trying to include too much.

Avoid:

  • full paragraphs
  • tiny feature lists
  • several competing calls to action
  • unreadable product screenshots
  • text touching the edges
  • oversized logos
  • vague decoration that does not support the message

The card does not need to close the sale. It needs to earn the next few seconds of attention.

Keep the important headline, logo, visual, and action away from the edges. Feeds and devices may crop or shrink the image.

Give important pages their own cards

Your homepage, feature pages, blog posts, launch pages, and changelog do not all have the same job.

A homepage card can explain the broad promise.

A feature card can focus on one outcome.

A blog card can lead with the lesson.

A launch card can make the announcement clear.

You can keep the same type, colors, logo placement, and general composition while changing the promise and visual for each important page.

This gives your website a reusable set of marketing assets. Whenever you share a specific URL, the card supports the message instead of dragging every conversation back to a generic homepage.

The DistributionOS before and after

DistributionOS now uses two purpose-built cards.

The homepage card leads with the pain:

Stop shipping apps nobody sees.

DistributionOS social preview card reading Stop shipping apps nobody sees

It gives a stranger a reason to care before asking them to understand the product.

The About page card leads with the outcome:

Know what shipped. Learn what worked.

DistributionOS workflow card reading Know what shipped. Learn what worked

It explains the loop behind the product and fits a page where someone is trying to understand how the idea works.

Both are more useful than the old screenshot because each card has a clear audience, message, visual, brand, and next step.

The only technical handoff most founders need

A practical master size is 1200 x 630 pixels.

It is a useful working default across common sharing surfaces. It is not a promise that every platform will crop the image in exactly the same way.

The implementation handoff is short:

  • use the card as the page's Open Graph image
  • ask X to use a large image card
  • make the image URL public
  • keep the important content away from the edges
  • paste the real page URL into the platforms where you plan to share it

If you use a coding agent, this is the easy part to delegate. Choosing the audience, promise, visual, and action is the part that needs your judgment.

One extra warning: social platforms cache previews. If an old image appears after an update, confirm that your live page uses the new card before redesigning it again.

A founder's card audit

Use these questions on every page you share regularly:

  • Who is this card for?
  • Can they understand the main promise in two seconds?
  • Does the visual help them recognize the problem or outcome?
  • Is the brand clear without taking over?
  • Is there one obvious reason to click?
  • Does the card still work when it is small?
  • Is important content protected from edge cropping?
  • Does this page deserve a more specific card than the homepage?
  • Have I pasted the real URL into the places my audience uses?

The test is simple: if someone sees the card without your accompanying post, can they tell whether the link is for them and why it is worth opening?

Sources

Evidence boundary

The 1200 x 630 recommendation is a cross-platform working default supported by current framework examples, LinkedIn's near-identical documented frame, and first-party testing. It is not represented as a permanent official requirement for every platform. Rendering and caching behavior should be rechecked before publication and periodically afterward.

Give your coding agent reusable distribution context.

Create an app record, build the Brain Doc, and let your agent ship marketing work with context instead of another blank prompt.

Connect your first app