Optimizing Social Previews: Open Graph & Twitter Images

When users share links to your SaaS product or blog posts on platforms like Twitter/X, LinkedIn, Slack, or Facebook, the platform displays a visual preview card. A shared link without an image or with a cropped preview card receives significantly lower click-through rates.

To ensure your shared links look professional and engaging, you must declare correct Open Graph (OG) and Twitter Card tags in your page HTML.

This guide explains how social previews work and how to optimize your card tags.


1. The Anatomy of a Preview Card

Social media cards are generated using metadata tags placed inside the <head> of your webpage:

  • og:title & og:description: The title and summary displayed below or on the preview card.
  • og:image: The absolute URL of the image displayed on the preview card.
  • twitter:card: Specifies the card format (e.g. summary_large_image for large header images or summary for small square thumbnails).

If these tags are missing, platforms will attempt to scrape random images or text from your page, leading to broken or irrelevant previews.


2. Best Practices for OG Images

To prevent cropping or pixelated cards:

  1. Resolution: Use an image with a resolution of 1200 x 630 pixels (a 1.91:1 aspect ratio). This matches the standard container dimensions for Facebook, LinkedIn, and Twitter/X large cards.
  2. Absolute URLs: The og:image URL must be absolute (e.g., https://yourdomain.com/og-image.jpg), not relative (/og-image.jpg), as external social scrapers cannot resolve relative links.
  3. Crawlable Image Paths: Verify that the image path is not blocked by your robots.txt file, allowing social media bots to download it.

3. Auditing Your Social Preview Previews

Instead of posting links manually to check previews, you can analyze your tags programmatically.

Use the OG Share Image Checker to inspect your URLs. It fetches the page HTML, verifies og:image and twitter:image presence, validates that the image URL resolves, and checks that the dimensions meet standard requirements for rich preview cards.

High-Converting Landing Pages: What 300+ Top Performers Have in Common

Research-backed patterns from 300+ high-converting landing pages: problem-first copy, audience language, one CTA, social proof placement, and the headline formula that converts.

Read more

The Mechanics of AI Retrieval: Exactly How ChatGPT and Perplexity Build SaaS Shortlists

A technical, evidence-based look at how RAG, source weighting, and third-party consensus decide which SaaS tools LLMs recommend — and what founders can actually influence.

Read more

Beyond GA4: How to Track and Attribute B2B Pipeline from ChatGPT and Perplexity

AI search traffic arrives via clean referral paths or stripped parameters, so it hides in Direct and Referral. A step-by-step setup for UTMs, GA4 regex channel grouping, and AI-aware analytics.

Read more

Conversation Demand vs SEO Content: What to Work on First

Two content calendars founders confuse - SEO territories from rankings vs conversation demand from communities. Order of operations by stage, with Trending Problems as upstream ideation.

Read more

How to Monitor Trending Problems to Validate Startup Ideas (2026)

Use the free Trending Problems feed and community research to spot rising pain points, validate ideas, and prioritize what to build - without surveys or guesswork.

Read more

LinkedIn Founder-Led Outreach in 2026: What Still Works (and What Doesn't)

Generic automated cold outreach is dead on LinkedIn, but founder-led brand plus warm outreach still converts. A realistic 2026 playbook with social-listening inputs.

Read more

Are you building a tool or platform in the GEO, AI marketing, or customer discovery space? Learn more about our editorial collaborations and sponsorship opportunities →

Find your next perfect customers

Turn this article's ideas into real conversations across 10+ communities.