Updated September 28, 2026
8 min read
Analytics

How to Analyze a Competitor's X Content Mix (Threads, Video, Links)

Counting Formats, Comparing Their Performance Inside the Account, and Reading the Shift

Short answer

A competitor's X content mix is the share of their posts in each format: text, image, video, link, thread, quote and reply. Count ninety days of posts by format, compare median views per format inside the account, and repeat next quarter. A move of ten points or more in any format is a deliberate change.

GB
Written byGabor Barta— Co-founder, Oppira

Gabor leads product and content at Oppira. He has spent over a decade building tools and writing about competitive intelligence, social media analytics, and growth marketing for B2B SaaS companies.

Published September 28, 2026

1. Which formats should I count on X?

Use seven buckets: text only, image, video, link, thread, quote post and reply. Keep reposts out, because they are someone else's content, and give polls their own row only if the competitor actually uses them.

Content mix
The distribution of an account's posts across formats and post types over a fixed period, expressed as percentages so that accounts with different posting volumes can be compared.
Also known as: Format mix, Post type mix
Format buckets for an X content mix and how to identify each one
BucketHow to recognize itWhat a rising share usually means
Text onlyNo media, no link cardOpinion, commentary, founder-style voice
ImageOne or more images attachedVisual product proof, screenshots, announcements
VideoNative video attachedInvestment in production; a bet on watch-time distribution
LinkExternal link, usually with a preview cardDriving traffic to a blog, landing page or launch
ThreadThe author replies to their own post to continue itEducation and authority building
Quote postTheir comment on top of another postReacting to the conversation rather than starting it
ReplyPosted in reply to another accountCommunity work, customer service or reply-led growth
Format buckets for an X content mix and how to identify each oneClassify a post by its dominant element. A thread that opens with a video goes in thread, and a link post with an image preview goes in link. Write the rule down so the next count uses it too.

2. How do I count a competitor's X content mix step by step?

Pick a ninety-day window, count total original posts with a dated search, use search filters to count links, images, media and replies, then classify threads, quotes and text posts by hand from the profile.

The examples use the invented handle @veltrix and one quarter of posts.

  1. Fix the window. Use ninety days ending on a fixed date. Shorter windows swing with one campaign; longer ones hide the shift you are trying to see.
  2. Count replies and originals separately. Search from:veltrix since:2026-06-01 until:2026-09-01 filter:replies for replies, and the same query with -filter:replies for everything else. The split between the two is the first finding.
  3. Count link and media posts with filters. Add filter:links, then filter:images, then filter:media to the originals query. Media minus images gives an approximate video count. Record each total.
  4. Classify threads, quotes and text by hand. Scroll the originals for the same window and tag each remaining post. Threads show the author continuing their own post in replies; quotes show another post embedded.
  5. Convert counts to percentages. Divide each bucket by total originals, and report replies as a separate ratio of replies per original post. Percentages make a busy account comparable with a quiet one.

3. How do I compare how each format performs?

Within the account only. Take the median views and median public engagement rate per format, then compare formats against each other for that competitor. Never compare one account video median with another account text median.

Formats distribute differently on X, so a blended account average moves when the mix moves, even if nothing about the content improved. Keeping a median per format is what lets you tell a real improvement from a format change.

The columns of a per-format performance sheet for one competitor over one quarter
ColumnWhat to recordWhy it matters
FormatOne row per bucket: text, image, video, link, thread, quoteKeeps formats from averaging into each other
Posts in windowThe count of original posts in that formatTells you whether a median is meaningful
Share of originalsPosts in format divided by all original postsThe mix itself, comparable across accounts
Median viewsMiddle view count of posts in that formatHow far X distributed that format for them
Median public engagement rateMiddle of per-post interactions divided by viewsHow well that format converted viewers
The columns of a per-format performance sheet for one competitor over one quarterOnly compute a median for a format with at least five posts in the window. Below that, write "too few" rather than a number that one post decides.

The link row is worth special attention. Many marketers believe posts with external links get less distribution on X. You do not have to take that on faith: compare the competitor median views for link posts with their non-link posts. Their own data tells you how the platform treats their links, and whether their link-heavy strategy is costing them reach.

4. What does a change in content mix tell me?

It tells you what they learned. A competitor moving from links to threads is chasing distribution; a jump in replies means community or support work; a jump in video means someone is funding production.

Common shifts and the usual explanations to test:

  • Links down, threads up: they saw link posts underperform and moved the content onto the platform, linking at the end of the thread instead.
  • Video up sharply: a production budget or a new hire. Check whether the videos are repurposed from another platform, which is cheaper and less committed.
  • Replies per original up: either a deliberate reply strategy on larger accounts, or a support load. Read a sample to tell which.
  • Quotes up, originals down: the team has less to say of its own and is commenting on the conversation instead. Often a sign of reduced resourcing.
  • Everything down: the channel has lost priority. Check whether posting moved to LinkedIn or Instagram in the same period.

Treat each explanation as a hypothesis until a second signal supports it. A job posting for a video editor, or the same videos appearing on Instagram Reels a day earlier, turns a guess into a finding.

5. How do I use this against my own plan?

Put your own mix next to two or three competitors in the same table. The useful output is a format that works inside their accounts and is missing or thin in yours, written as a specific test.

If Veltrix and Sondera both see their threads beat their own median while you post almost none, the test is concrete: publish one thread a week for six weeks and compare its median views with your other formats. That is a plan item. "We should do more threads" is not.

Oppira tracks competitor X posts daily and charts posting cadence and format per rival and head to head, next to the same view from Facebook, Instagram and LinkedIn. The quarterly count becomes a comparison you read rather than one you rebuild, and cross-platform shifts show up side by side.

Key Takeaways

Seven buckets, reposts excluded

Text, image, video, link, thread, quote and reply cover almost every competitor post on X.

Replies are their own ratio

Report replies per original post separately, because they measure community work, not content output.

Use filters, then check by hand

Search filters count links, media and replies quickly, but results are approximate and need a spot-check.

Compare formats inside one account

A median per format keeps a mix change from looking like a content improvement.

Test the link question on their data

Compare median views of their link and non-link posts instead of relying on received wisdom.

A shift needs a second signal

Confirm a mix change with hiring, cross-platform reuse or a launch before reporting a reason.

Frequently Asked Questions

Sources

  1. How to use advanced search X Help Center, September 2026.Official reference for filtering search results by account, date and content type.
  2. View counts X Help Center, September 2026.
Oppira

Turn reading into a reaction

Oppira watches your competitors, keeps your playbook current, and drafts the response. See your market clearly by tomorrow.