What’s new in Reptor

Established falsehoods

Not every negative mention is untrue, and not every untruth is negative. Reptor now tests what is written and said about you against the knowledge you have approved, and flags it the moment the two contradict each other. Contradictions between AI platforms are picked up as well: if one platform claims something different from another, that is a signal in itself. You decide which side is right, and each side then moves to where it belongs. What holds up becomes knowledge, what does not lands among the established falsehoods and counts in everything analysed afterwards. The check runs across all your sources, reviews included.

AI Overviews added

Google increasingly puts an AI answer above the search results, and for many people that is the only thing they read. Reptor now brings those AI Overviews in as their own AI platform, measurable per question, along with the sources cited underneath. Those sources link through to their own page, so you see exactly which sites shape what Google tells people about you.

AI visibility: who stands next to you

Knowing whether you are mentioned is one thing. Knowing who is mentioned alongside you is the real story. The new AI visibility section shows which brands and people appear in AI answers about your topics, how often you are among them, where you are missing while someone else is there, and which themes are at play. Reptor keeps the distinction between questions where you want to be named and questions where absence is the goal, because with a case or an affair, not being named is the best outcome.

Your knowledge profile learns, with your approval

Everything Reptor knows about you now sits in one place, per entity, and that profile is the source all analyses reason from. It fills itself in from your own website and from what is published about you, but nothing becomes active without your approval: proposals come in for review and you approve or reject them. Your corrections apply immediately to the analyses that follow, so the profile sharpens the more you use it. And because the knowledge belongs to the entity rather than the monitor, a parent company and a subsidiary keep their own details.

The questions people actually ask

The AI questions we measure your visibility on used to come from what we could think of. They now start from what is genuinely being searched about you: the search suggestions, follow-up questions and related searches Reptor already collects for you. Every search term shows the AI question that belongs to it, so it is clear what exactly is being measured.

The link graph

Who links to you, and where do you link out to? Reptor now records this while fetching a page, so it costs no extra measurement. Per receiving page you see how many links come in and from how many different domains. Per outlinking domain you see which source page points at which target page, with the anchor text and whether the link is followed. Your own channels are recognised separately, and for a mention you also see how often it was shared on social and in which direction. The whole graph exports in one click.

Trustpilot linked automatically

With a common company name, the right Trustpilot profile is not obvious to find. Reptor now recognises the profile itself and verifies it against your domain and your website before linking it. If a profile exists but carries no reviews yet, you hear that too, because an empty profile is a reputation fact in its own right.

Why this verdict

For every mention you can now see why Reptor found it relevant or not, and why the sentiment is what it is. The reasoning sits with the verdict, and the buttons explain what your choice does. More importantly: when you correct a verdict, that correction counts in the analyses that follow. Borderline cases are automatically escalated to a stronger model, so the hard ones get more attention than the easy ones.

Images and video become text

A post about you can live entirely in a photo or a video, leaving nothing in the text. Reptor now turns that imagery into text: social posts get a description of what is shown, and videos on TikTok and Instagram are transcribed, just as podcasts already were. Whatever is said or shown on screen becomes searchable and gets a sentiment score like anything else.

Under the hood: sturdier, sharper, easier to read

Alongside the new features, a great deal of work went into precision, stability and overview in this period.

Channels catch more. Bluesky posts arrive with the shared link card, quotes and alt texts intact. Instagram and X profiles now produce mentions themselves. Comments under Facebook posts are counted by type and weighed in. For audio the matcher scales with the length of the search term, person names match exactly and quotation marks count as a single phrase again, which makes the hits noticeably more precise. Spelling variants can be managed per search term.

Faster to read. The Intelligence tab, the Search tab and the detail page have been reorganised: fixed columns, real platform logos, a legible position graph you can click per day to see the results of that day, expandable AI answers and negative results grouped together. Visibility is shown as a percentage instead of a count, themes link through, and the whole dashboard now runs on the latest house style. Site type and publication type are visible and correctable, and the full list of categories is available rather than the first few.

Everything in your language. Summaries, reports and the interface consistently follow the language of your account, regardless of the language searched in.

Solid under load. Transcription runs on heavier hardware in batches and keeps up with peak moments. If an AI platform or search source is briefly unresponsive, Reptor takes an alternative route and the scan simply continues, with page numbers and positions that hold up on that route too. Alerts for podcasts run through the same delivery chain as the rest, and no alert goes out before relevance has been assessed. Publication date and modification date are kept apart, so an old article that was touched again does not resurface at the top of your timeline. Duplicate journalists have been merged and media groups filled in, so Top Voices counts nobody twice. And the assistant sticks strictly to your monitoring data, with a feedback button on every answer.

Chat with your monitoring

Your monitoring now holds thousands of mentions, sources and signals. From now on you simply ask it a question: "What was the most negative article this month?", "Which journalists wrote about us?", "Summarize last week's social discussion." The built-in assistant searches your own monitoring data and answers in plain language, with the underlying mentions attached so you can click straight through. No more digging through filters and tabs: you ask, Reptor looks it up for you.

Comments under social posts now count

A post about your brand can read as neutral while the comments underneath tell a very different story. Reptor now also reads the comments under posts on Facebook, TikTok, LinkedIn and more, and factors them into the sentiment analysis. So you see not just what is posted about you, but how the audience responds, with the most relevant comments right next to the mention.

Share counter: how often content about you is shared

How often an article or post about you gets shared says more about its reach than the piece itself. Reptor now counts, per mention, how often it is shared across social platforms. At a glance you see which coverage is really going around, so you can prioritize your response on what is actually spreading.

AI-cited sources are now tracked

When an AI platform answers a question about your company, it draws on sources. Reptor now automatically extracts those cited sources from the AI answers and adds them to your monitoring, so you see exactly which pages shape your AI reputation and can track and influence them yourself. Tracking and redirect links are filtered out automatically, leaving only the real sources. A concrete extension to your GEO strategy.

Under the hood: more accurate, more stable, less noise

Alongside new features, a lot of work went into reliability. The same social post now appears only once in your overview, no matter which link it came in through. Borderline relevance and sentiment cases are automatically escalated to a stronger AI model for a sharper verdict. Transcription has been decoupled from the recording so no fragments drop under load. Pages we are not allowed to visit now get a fitting title. And when an AI platform is briefly unresponsive, Reptor falls back to an alternative source instead of letting the scan fail. In short: the same Reptor, tighter.

Language filter: you decide which languages count

Set the active languages per monitor and Reptor shows only content in those languages, consistently across your overviews, dashboards, trends, monthly reports and alerts. On the Media and Social tabs you can also filter by language yourself. Nothing is thrown away: content in other languages keeps its language label and reappears the moment you activate that language. Especially valuable for organizations with international namesakes: foreign coverage about a company that happens to share your name no longer clouds your sentiment picture, while international brands simply activate multiple languages.

Namesake detection: automatic disambiguation

Companies and people with a common name used to see mentions of namesakes slip into their monitoring. Reptor now runs an AI namesake analysis right after the first scan: it clusters all found mentions per real company or person, shows which namesakes exist, and suggests exclusions and search refinements you can apply in one click. Social posts are checked on the poster's identity as well, and the noise filter has been sharpened further. The result: drastically fewer false positives and a sentiment picture that is really about you.

Video transcription for social videos

A mention in a TikTok or X video used to be invisible: spoken words appear nowhere as text. Reptor now automatically transcribes social videos about your brand, so what is actually said in them becomes searchable and gets a sentiment score, just like written content. After podcasts this closes the last audiovisual blind spot in your monitoring: video.

Boolean search terms (AND, OR, NOT)

Search terms now support boolean operators: AND, OR, NOT, quotes for exact phrases and exclusions. That makes monitoring far more precise for brands with generic names or multiple spellings; think of an exact phrase like "ACME" instead of every page containing those letters, or excluding an industry that shares your name. Your operators carry through to all sources and the relevance filter respects your exclusions too.

Microsoft Copilot monitoring: now 7 AI platforms

Reptor now also monitors Microsoft Copilot. With Copilot built into Windows, Microsoft 365 and Bing, more and more people get answers about your company straight from Copilot, so what it says about your brand matters just as much as ChatGPT or Gemini. Reptor now covers all major AI platforms: ChatGPT, Gemini, Perplexity, Grok, Google AI Mode, Claude and Copilot, giving you full visibility into how AI portrays your brand.

Podcast monitoring

Reptor now also monitors Dutch podcasts. Every new episode is automatically transcribed and searched for your brand, names and topics, with a sentiment score and a deeplink to the exact moment in the episode. Podcasts were invisible in reputation monitoring until now: spoken mentions appear nowhere as text, while hours are spent talking about companies, executives and industries. Reptor now covers the audio channel that stays invisible in text.

Send pitches from Reptor + build your own clippings

Agencies can now send press releases and pitches directly from Reptor to their journalist list, with automatic open and click tracking. Follow-up publications that Reptor picks up are linked back to the original pitch, so per campaign you can see which pitches land and which outlets pick them up — a measurable layer on top of monitoring. Clients can also now build their own clipping reports from the dashboard and mark URLs for sentiment or relevance; that feedback feeds the self-learning noise filter directly, making the system more personal to your brand. Finally, journalist and media profiles now open as full pages instead of modals, with more history and contact details at hand.

Monthly reports 2.0: stories, top voices and content gap

Monthly reports got a major upgrade. A new Stories section clusters related articles into a single running narrative (news peak, event, ongoing, background theme), so you no longer see the same event five times in one report. Top Voices shows the most relevant voices per platform with sentiment bars and reach. Content Gap Analysis reveals which publications are writing about your competitors but not about you — concrete leads for your PR work.

Clipping reports with shareable AI summaries

Bundle articles, posts, and reviews from the dashboard into a clipping report: nine modular sections (sentiment, journalists, social, AI citations, reviews, theme), PDF and Excel export, email delivery, and a shareable public link. Generate an AI summary per report, then ask follow-up questions in a running conversation. Every version gets its own share link so older snapshots stay preserved and you can send a tailored summary to each stakeholder.

Backsearch + LLM source analytics

Reptor monitors continuously, but what if you add a term, person, or crisis later? With Backsearch you can search retroactively across every monitored channel over a chosen period. For audio, Reptor keeps all transcripts in a rolling 90-day window, so spoken mentions stay searchable after the fact. The LLM tab also got a source analytics dashboard that shows which domains and URLs are cited most often by AI platforms for your search terms — with drilldown per domain, search, and history. Directly usable for your content and GEO strategy.

Claude AI monitoring — 6 AI platforms

Reptor now monitors Anthropic’s Claude as a sixth AI search engine. As more people use AI assistants instead of Google, what Claude says about your company matters just as much as what ChatGPT or Gemini says. Reptor now covers all major AI platforms: ChatGPT, Gemini, Perplexity, Grok, Google AI Mode, and Claude — giving you full visibility into how AI portrays your brand.

Bluesky monitoring + expanded review platforms

Added Bluesky as a monitored social platform via the public AT Protocol API. Review monitoring expanded with Google Play, App Store, and Klantenvertellen. New Intelligence dashboard with journalist tracking, media outlet analysis, and competitive benchmarking.

Google AI Mode monitoring

Added Google’s new AI Mode as a fifth AI platform. Reptor now monitors the AI-generated answers that appear directly in Google search results.

Social media monitoring expansion

Added TikTok and Reddit monitoring. Instagram and Facebook monitoring improved with deeper content analysis.

Enhanced alert management

Improved alert workflow with bulk actions, priority filtering, and sub-platform selection for Google Search vs. Google Maps.

Self-learning relevance system

Reptor now learns from your feedback. Mark mentions as relevant or irrelevant, and the system adapts its filtering for future scans.

Complete dashboard redesign

New position tracking charts, improved SERP feature cards, and a unified media timeline. PDF export with custom headers.