TL;DR
- A news outlet isn't a brand. It publishes more often, depends on the news cycle, lives off discovery and competes with creators and algorithms, not just other publishers.
- That's why measuring like a brand (likes, impressions, followers) falls short. The question to ask is: which content wins attention, conversation, visits, loyalty or ground against the competition?
- Five challenges: tricky attribution, platform volatility, fragmented logics (X ≠ TikTok), adapting editorial language to platform language and separating reach from quality of attention.
- X is best measured as a network of speed and conversation. TikTok, as a network of discovery and format. And social platforms are increasingly news search engines.
- Welov.io helps newsrooms read all of this from a single view, comparing with competitors using public data, without resorting to mass social listening.
Analyzing social media in news outlets means measuring the editorial content published on social platforms (your own and your competitors') to understand what works, why, against which references and with what impact on the audience and the business. It's not the same as measuring a brand's social media.
For a news outlet, the problem isn't publishing a lot. It's understanding why some pieces work and others don't, how each publisher compares to its competition and how the social team's value gets defended in front of management.
News consumption has shifted strongly toward social platforms, video and intermediary platforms. Creators are now competing for the same attention. And the drop in social referrals forces better measurement: not just reach, but also attention, loyalty and business impact. The 2026 Reuters Institute data points in that direction: more video, more fragmented consumption, more pressure to rethink distribution.
This article isn't an abstract guide on "how to measure social media". It's a guide on how news outlets should measure it, given that their reality is different from a brand's: high frequency, dependence on news cycles, the need to react in real time, difficulty attributing traffic and a competitive landscape where they no longer compete only against other publishers but also against creators, native accounts and algorithmic recommendations.
THE FRAMING SHIFT: FOUR QUESTIONS, NOT FIVE THOUSAND METRICS
Social analysis in news outlets can't stop at "likes, impressions and followers". At a minimum, it has to answer four questions:
- Which content wins attention? Who watches, how much and with what quality.
- Which content wins conversation? Replies, quotes, comments, shares.
- Which content wins visits or loyalty? Site clicks, time engaged, return rate.
- Which content wins ground against the competition? By topic, by format, by time slot.
Each question points to a different layer of editorial work. If the team doesn't separate them, everything blurs into the same dashboard and the analysis loses force. If they do separate them, useful reading appears; the kind that supports decisions, not the kind that fills a spreadsheet.
FIVE CHALLENGES WHEN A NEWS OUTLET MEASURES ITS SOCIAL MEDIA
1. Attribution is tricky
Many people consume news almost accidentally, inside the feed, without clearly remembering which publisher produced the original piece. The complexity of the online ecosystem and the weight of social platforms, aggregators and search engines make it very hard to correctly attribute the source of a story. When consumption is passive, memory of the source degrades.
For a news outlet, this complicates both brand analysis and a real read on each platform's contribution. The "traffic from X" metric doesn't capture "people who saw the news thanks to us but came in through another door".
2. The consumption platform has changed
Reuters Institute reports referral drops to news sites of 43% from Facebook and 46% from X over the last three years. At the same time, many publishers are dedicating fewer resources to X, even though the platform still holds news weight in some markets.
There's a real mismatch between "where conversation still happens" and "where it no longer pays off to distribute the way we used to". The team has to decide what to measure and over what horizon, knowing that the right answer today might not be the right answer in six months.
3. X and TikTok aren't measured the same way
X works more as a news destination and breaking-news platform. TikTok is a network of discovery, algorithmic recommendation, native video and creators. There's no point in measuring both the same way or expecting the same return.
The same Reuters Institute report shows that X holds up as a news network in some markets, while TikTok grows as a news access channel and concentrates less attention for traditional outlets within the social ecosystem.
4. Adapting journalism to platform language without diluting it
Recent research on journalistic TikTok suggests that many pieces are now standalone news products designed for the platform's aesthetic and logic, not clips of the web piece. News outlets adopt visual and hashtag resources well. They adopt sound resources much less.
The challenge isn't only to measure better. It's to understand which parts of the native language the team is mastering and which it isn't. And that only becomes visible when you cross performance with qualitative format analysis, something raw metrics don't deliver.
5. Reach isn't the same as quality of attention
On X, many people see breaking news and stories from outlets or journalists, but they also perceive a lot of inaccurate information. On TikTok, according to Pew Research Center, half of users get news there regularly, even though less than 1% of the accounts they follow belong to institutional sources. Access is indirect and unintentional.
For news outlets, this means it's not enough to ask "did they see us?". You also have to ask "in what context did they discover us?" and "what kind of competition did we share that space with?".
THE METRICS THAT ACTUALLY MATTER
- Whether your content enters circulation. This includes impressions, reach, plays and publishing volume.
- How the audience responds. Not all interactions count the same. On X, the mix of likes, reposts, quotes and replies changes the type of circulation. On TikTok, comments and shares carry particular weight, because they speak to forwarding and conversation.
- What happens after social impact. URL clicks, profile clicks, traffic to the site, visit depth, time engaged, return frequency. Traffic by source, time engaged, return frequency and direct traffic are central variables for publishers. The question isn't only whether the post got engagement. It's whether the visit it generated had quality.
- The competitive context. Knowing that a piece got 50,000 interactions says nothing on its own. The question is whether that figure is high or low compared to other publishers, on the same topic, in another format and in another time slot.
X AND TIKTOK: HOW THEY'RE USED IN THE NEWSROOM AND HOW THEY SHOULD BE MEASURED
X: a network of speed, conversation and activation
On X, news outlets still find a network that's especially useful for breaking news, real-time coverage, article circulation and conversation with journalists and politically engaged communities. Pew Research Center shows that half of X users get news there regularly, that 65% use the platform at least partly to stay informed, that 75% see breaking news in real time and that 80% of those who consume news on X get it from outlets or journalists.
What to measure:
- Impressions and interactions by topic or event, with a focus on comments.
- Reposts and quotes (qualified circulation vs. passive likes).
- URL clicks and profile clicks (with the caveat that X considers these metrics private to your own posts).
- Publishing pace around events, not just aggregate engagement; especially relevant when comparing with the competition.
TikTok: a network of discovery, format and affinity
TikTok runs on a different logic. News outlets produce native, standalone pieces adapted to the platform's visual logic. And creators compete intensely with journalistic brands. Reuters Institute finds that traditional outlets' attention is weaker on TikTok than on other platforms and that a large share of people receive news there without actively following any outlet.
In practice, news outlets use TikTok more for discovery, explainers, journalist personalities, visual formats and connection with younger audiences than for classic referral.
What to measure:
- Plays, ER per follower, shares.
- Pace by time slot.
- Qualitative reading of format, angle and tone.
The difference that changes the dashboard
X is best measured as a network of speed, conversation and activation. TikTok is best measured as a network of discovery, format and affinity.
On X, what matters is whether the content wins informational conversation and useful clicks. On TikTok, what matters is whether the content finds the right native angle, fits into the platform's codes and builds habit or awareness among people who don't even follow the outlet.
If the team applies the same dashboard to both, the reading won't be optimal.
SOCIAL PLATFORMS AS NEWS SEARCH ENGINES
The idea of "social platforms as search engines" is no longer a metaphor, especially on TikTok. The platform maintains a Discover and search space where users explore videos, hashtags, creators and sponsored content. It personalizes results based on prior searches and views, and tries to highlight reliable sources at the top. It also offers Creator Search Insights to show topics people are searching for and to track how publications perform in search results.
From the consumption side, the most common way to come across news on social platforms is still seeing trending stories, comments and links shared by followed accounts, especially on text platforms like X. On video platforms like TikTok, one in three people who use the platform for news say they watch "videos about current stories". Search and visual discovery are now part of the news experience.
For a newsroom, this means changing the question. It's not "which post got the most views?". It's "which content works as a quick answer?" and/or "which content works as search intent?".
Useful analysis combines topic or implicit query, keywords in copy, format, time, comparison with competition, URL clicks or referred visits where they exist, and on-site behavior if the piece ends up driving traffic to the site.
The good news: you don't need to enter social listening to understand whether a network is working as a news search engine for an outlet. It's enough to analyze the performance of published content, see which topics and phrasings capture discovery, compare with competitors and cross that signal with traffic and attention when the data is available.
HOW TO LAND THIS WITH WELOV.IO
Welov.io isn't a tool for mass conversation listening. It's a layer of own and competitive content analysis. And that difference matters, because for a newsroom it's exactly what's missing: reading published content better, not chasing every mention.
Welov.io's content views allow you to combine publications from different profiles and platforms in the same grid, sort by the metric that matters, work by date ranges and turn that selection into custom reports. AI Insights adds a qualitative reading that helps explain performance, detect topic trends, gaps versus competitors and territories with stronger affinity. The step from data to editorial judgment.
Four concrete advantages for a news outlet:
- Measuring competition on TikTok and X with public data. When comparing publishers is part of the daily job, ditching gut estimates changes the internal conversation.
- Having enough history to spot patterns. On TikTok, Welov.io can recover up to two years of data prior to the connection for own profiles and, for competition, by default it works with the last 50 videos and on X with the last 100 tweets or from the start of the previous month, whichever is more recent.
- Working at volume. Grids, reports and exports to analyze many publications without drowning in Excel: the difference between reviewing 30 pieces and reviewing 300.
- Reading copy and narrative patterns with AI. Tone, style and content construction supported by real performance history and custom prompts.
A real case. Axel Springer Spain (publishing) recovered 30 hours per month with automated reporting in Welov.io, and recorded a +16% increase in social traffic versus the previous year. The difference wasn't adding another dashboard. It was no longer building it by hand and using that time to read better what was already happening.
If we had to compress it into one idea: for news outlets, Welov.io doesn't add value by piling up dashboards, it adds value by helping answer questions that native tools and spreadsheets don't answer well at scale.
Some questions Welov.io can answer for a news outlet
- Which angle works
- Which tone drives more interaction
- Which competitor is gaining ground
- Which format generates more impact
- Which pattern deserves to become an editorial guideline
It's worth remembering that, even with good analytics, competitive analysis on social is conditioned by the increasingly intermediated nature of news consumption. That's why an approach like this gains weight: less obsession with isolated vanity metrics, more combined reading of format, topic, tone, competition, traffic and attention.
A news outlet measures more, measures faster, measures against a different competitive set and measures to answer to a newsroom.
If your team is building reports in Excel every month, eyeballing comparisons against three publishers or trying to justify an editorial decision without data behind each format, there's a problem. Once you've changed the framework, a layer like Welov.io helps you do it at scale: own content, competition and qualitative AI in one view, without touching Excel.
Do you work in a newsroom and want to see how your analysis would look with this? Try Welov.io free for 14 days.
