Monitors can now track Meta AI, the assistant at meta.ai, alongside ChatGPT, Claude, Gemini, Perplexity, and Google's AI answers. Meta AI searches the web before it answers, so you get its citations and query fanouts like any other live-search model.
A monitor can now run its prompts several times a day instead of once. The tracking frequency picker has a new Several times a day group with every 12 hours, every 6 hours, and every hour.
Socials now covers Instagram next to Reddit, YouTube, and LinkedIn. See which Instagram accounts, reels, and posts AI answers cite for your prompts, with their likes, comments, plays, and follower counts.
The Agent Chat panel now opens on many more pages and answers from what's on screen, with the filters and date range you have set.
Site health has a map of how your pages link to each other, and every page shows how it got into your audit.
A few smaller changes across the platform:
A round of fixes across the platform to keep things running smoothly:
Topics moved out of the Prompts tabs into its own page under Trends, and it shows how your brand does in each topic for the dates you pick.
A new Persona insights page shows how your visibility, position, and sentiment change per persona. Each persona gets a radar chart against the general audience, the answers that ran without a persona, so you can see which audience AI recommends you to least.
Onboarding has a live preview of the dashboard next to the form. Your brand, competitors, and prompts fill in as you enter them, and the preview switches pages as you go through the steps.
A few smaller changes across the platform:
A round of fixes across the platform to keep things running smoothly:
Filter the Shopping overview by AI model and date range to see which products each model recommends, and how that changed over time.
ChatGPT stopped sending a referrer on October 8, so visits from ChatGPT started showing up as direct traffic. ChatGPT has acknowledged the bug and still adds `utm_source=chatgpt.com` to its links. Turn on Include ChatGPT UTM referrers in Visitor Analytics to count those visits as ChatGPT again.
The AI chat panel from the content editor now opens next to your monitors. On a monitor's overview, prompts, responses, and citations tabs, ask about what's on screen and the agent answers from that page, with the filters and date range you have set.
Start a custom dashboard from a template instead of an empty page, or ask Agent Chat to build one for you.
Every Tuesday you get an email with the latest site audit for your projects: the health score, how many issues were fixed since the last audit, and the top issues that AI crawlers may trip over, with a link to Site Health for each project. You can turn it off under email preferences in your user settings.
The content agent no longer needs a tracked prompt. Under **What to write about**, choose **Describe a topic** to write about any subject, or **Target a tracked prompt** to use its query fanouts and the competitor pages AI engines already cite. You can also start an article straight from the places where you spot a gap.
The competitor heatmap is now a general heatmap. Put brands, models, topics, or countries on either axis, and color the cells by visibility, sentiment, position, share of voice, or win rate.
The content editor has an AI chat panel next to the draft. Open it with the AI chat button in the header or ⌘J, and ask it to rewrite, extend, or check the article you're working on. The agent reads the whole draft, long ones section by section, and edits it in place, so you don't have to copy text back and forth.
A few smaller changes across the platform:
A round of fixes across the platform to keep things running smoothly:
The Brand Book has a new Analysis instructions card. Write plain-language rules, up to 1,000 characters, for what should count as a mention of your brand and your competitors. Promptwatch applies them when it analyzes new responses, so visibility, sentiment, share of voice, and competitor rankings follow your rules. Responses that were already analyzed keep their results.
The REST v2 API and MCP server accept date ranges of up to 92 days, so a full three-month window from Looker or another BI tool no longer gets rejected. Agencies can also create pitch projects through the API and MCP. They count toward your pitch project allowance, not your regular projects.
Codex and Claude Code are now available as models for every customer. Add them to a monitor to see whether coding agents recommend your product when developers ask them to pick a tool, library, or service. Both search the web while they answer, and their query fanouts are collected like the other models.
Custom dashboards can now be exported and scheduled as PDF reports. The report contains the widgets you put on the dashboard, in the same order, and is titled with the dashboard's name. A schedule follows its dashboard, so if you change the widgets or rename the dashboard, the next report picks that up.
The Content Agent now looks for relevant pages on your site before it starts writing, so links to them are part of the article from the first draft. Pages that get many Search Console impressions but few clicks are suggested first. It also stopped linking your homepage every time the brand name comes up.
Organization settings has a new Audit log. It lists every change made in your organization, who or what made it, and when, whether it came from the dashboard, the API, MCP clients, or Agent Chat. Filter by person, source, or outcome, or search by action. The audit log is available to organization owners on Business plans and higher.
Add the Reddit, YouTube, and LinkedIn accounts your brand runs on the new Social channels page. Then turn on My channels on any of those platforms under Sources to see only the citations of your own subreddits, channels, pages, and profiles.
Agent Chat now draws charts when an answer is easier to read as one, including answers from its research agents. Period comparisons are overlaid in one chart, and colors match their meaning, like your brand versus competitors. When a request is unclear, the agent asks a short multiple-choice question before it starts, and at the end of a turn it can suggest next steps you pick with one click.
Custom dashboards are no longer a copy of the Overview. Add the widgets you want, such as brand visibility, sentiment, visibility by country, the competitor heatmap, top cited domains and URLs, and top YouTube videos and Reddit posts, then drag them into order. New dashboards start empty and open in edit mode. You can also find any saved dashboard from the command menu.
Content gap recommendations now show whether the Content Agent has picked them up, reviewed and held them back (with the reason), or queued them for its next planning run. The same status shows per prompt in the content gap table. If a draft already exists, whether you wrote it or the agent did, View document is the main action, so you're not asked to create the same piece twice.
Three new data reports on what AI engines cited in August 2026:
The Topics tab on the prompts page has two new views next to the default table. By Competitor is a heatmap of visibility per topic for your brand and your competitors. By Position ranks the brands per topic. Both follow the date picker and can be searched and filtered by tag.
Agency plans now include visitor analytics: 500K monthly page hits on Kick-off, 2M on Growth, and 15M on Scale. You can see how much traffic AI search sends to each client's site, next to their AI visibility.
Save the Overview with your own filters as a dashboard. Each saved dashboard gets its own tab and keeps its date range, models, prompt types, topics, and monitors, so the view you check every week is one click away. Share a dashboard with the team or keep it personal.
A few smaller changes across the platform:
A round of fixes across the platform to keep things running smoothly:
When your organization reaches 85% of its agent credits, articles, or prompt responses, Promptwatch opens the usage overview once so you see it coming. Credits and articles can be topped up with an add-on right there; for responses you get a plan upgrade. Only organization owners see these alerts.
Creating a monitor now suggests a set of starting prompts before you save, so a new monitor starts collecting answers right away instead of sitting empty. New projects also get two extra monitors after onboarding, each aimed at a different country or city, so you see how your visibility changes between markets from day one.
Prompt Explorer has a new Ranking keywords tab. Enter a page or domain and Promptwatch looks up the Google keywords it ranks for, then turns them into prompt suggestions you can add to a monitor. It sits next to the Search Console, People also asked, seed query, and website page sources.
The REST v2 API and MCP server picked up several new endpoints. Google Search Console and Bing Webmaster page and trend data are available per engine, LinkedIn citations sit next to Reddit and YouTube, and organization-level keys can read a usage report with plan totals and a per-project breakdown. Agencies can also create a project, enrich it, and generate up to three monitors with prompts in a single call, and the MCP createMonitor tool accepts the same prompt options as REST.
Select several pieces on the Content Agent page and accept, decline, or publish them in one go. Failed pieces can now be deleted, from the bulk toolbar or the detail sheet, with an option to delete the linked draft as well. Content is also written in the language of the target prompt or monitor, so a Dutch prompt gets a Dutch article without changing any settings.
The prompts page can now rank brands per topic. Each topic row shows the top 10 brands by visibility over the last 30 days, with your own brand highlighted, so you can see who owns each topic at a glance. The By rank view can be exported, and the prompts list has a model filter that also scopes the prompt type and intent bars.
New personas get a generated portrait that shows up in the Brand Book and on response cards, with a letter placeholder while the image is being made. Personas also have an optional gender field.
Every export dialog now shows a 10-row preview next to its filters, so you can check the output before you download it. Most export types also let you pick which columns to include, and the responses export was redesigned around the new layout.
The onboarding checklist in the sidebar now opens a setup guide. Tasks are grouped into steps, and adding a sitemap, inviting teammates, and connecting Google Search Console can be done right inside the guide. Items that don't apply to you can be dismissed.
A failed payment no longer pauses your data. While a subscription is past due, prompt responses, crawler logs, visitor analytics, and Search Console and Bing syncs keep running, so there are no gaps in your charts once the invoice is paid. Chat, reports, publishing, and the Content Agent stay paused until then.
Connecting Webflow or Framer now suggests how each CMS field should be filled, and Promptwatch can use AI to fill fields that don't map to anything in the article, like categories, references, or custom text. Field help text from your CMS is shown in the mapping, and AI-filled fields are marked so you know which values were generated.
Visitor and crawler analytics can now stack include and exclude rules on page paths, matched partially or exactly. Every chart, table, and crawler export on the page uses the same filter, so you can, for example, look at blog traffic without the tag pages. GEO 360 Topic insights also has a page URL filter that narrows the landscape, top topic cards, and topic sheet to matching pages.
WordPress is now an option in the crawler log setup. Install the Promptwatch plugin, paste an API key, and AI crawler visits start coming in without touching your server or CDN config.
Essential and Professional plans can now run monitors in any supported country, the same as Business and agency plans. Explore stays tied to the project's country.
A 12-month snapshot of IP-verified AI crawler traffic, broken down by industry. Pick an industry to see which crawlers visit most and compare your own mix to peers, grouped by crawler or by provider.
Check whether AI agents can work with your site. Enter a domain and the tool shows which AI crawlers can reach it and which agent standards it supports, including llms.txt, MCP, A2A, OAuth discovery, and Content Signals. Every result gets a shareable page.
The 360 Dashboard has a new Topic insights tab. Every page on your site is grouped into topics, and each topic shows how many pages it covers and the average AI crawls, citations, and clicks those pages get. A bubble chart lays out the whole landscape, so you can see at a glance which topics AI engines pick up and which ones sit idle.
Two new tabs on the Sentiment page show how AI talks about you next to your competitors. By topic lays out sentiment per brand for every topic you track. By competitor gives one row per brand with the sentiment split across positive, neutral, and negative, so you can see where a competitor wins the tone and you don't.
Socials now covers LinkedIn alongside Reddit and YouTube. See which LinkedIn posts and authors AI answers cite for your prompts, how often, and how that changes over time. Filter by prompt, topic, or tag, and export the table like any other citation view.
The REST v2 API and MCP server now return the same share of voice you see on the dashboard, on the responses summary and brand visibility endpoints. Response analytics and query fan-outs accept a models filter, and citation rank comes back as a 1-based answer position. Agent Chat can also search the Promptwatch docs, so questions about how a feature works get answered from the documentation instead of a guess.
A round of fixes across the platform to keep things running smoothly:
The prompts page has a new By topic view. Each topic rolls up its prompts with average visibility, average sentiment, prompt count, and analyzed responses, and expands to show the prompts underneath. Filters work the same way as the default table, so a topic row always matches what you'd see when filtering the list.
The action board is no longer limited to what Promptwatch detects. Create your own actions with a title, description, severity, and status, edit them later, and dismiss accepted items so they leave the board but stay under dismissed actions. @-mention the agent in an action comment and it replies in the thread.
The 360 Dashboard gets a Search insights tab that puts Google Search Console and Bing Webmaster performance next to your AI data. See clicks, impressions, and top queries per page, split branded from non-branded queries, and open a page to compare its search traffic with AI crawls and citations.
Articles now have a separate meta title. The editor labels the old title as H1 and adds a meta title field that falls back to the H1 when empty. Webflow and Framer field mappings can point at it, and it comes back in the REST v2 and MCP content responses. Page Tracker can filter to pages that came out of the Content Agent, and the competitor picker in the content brief shows how many unique citations each competitor has for the selected prompt.
Organizations on Essential or Professional can submit a G2 review from the sidebar. Once approved, you get 2,000 extra agent credits every month for Agent Chat and the Content Agent.
Enter a website and watch a live scan of how AI talks about the brand. The report reads robots.txt, extracts what AI thinks you do, then asks real engines the questions buyers ask. In a couple of minutes you get scores for visibility, mentions, position, citations, and whether AI crawlers can actually read the site.
Ask Agent Chat to assign an action or leave a comment, and it writes those updates as itself. They show up on the action timeline so the rest of the team can see what the agent changed.
A round of fixes across the platform to keep things running smoothly:
WordPress publishing is open to every organization, not just the waitlist. Connect the official plugin, set categories and an author on the connection, and Content Agent articles publish with those defaults.
The main dashboard now has a world map of brand visibility. See which countries you show up in, which ones you haven't started tracking yet, and click a country to create a monitor for it.
Visibility charts now separate the period average from the latest score. Switch between Visibility over time, Visibility, Share of Voice, and Position. Prompt detail sheets get the same switcher even when the metric tabs are hidden.
Move is now Copy. Prompts stay on the source monitor, and copies land on the monitors you pick. Locale mismatches are blocked, and the dialog stops you if the copy would go over your prompt limit.
Articles published to a localized Webflow site now land in the locale that matches the article's language. New items are created across your enabled locales, and later updates don't overwrite translations you already made in Webflow.
Finished drafts include a "How this content was created" card that shows the inputs, research, and internal links behind the article. Internal links prefer pages that already perform in Google Search Console, leftover generation budget can schedule into the next period, and the featured image lives in the editor sidebar.
Organization owners can cap active prompts and prompt responses per project from organization settings. Usage, the upgrade banner, and tracking projections follow those caps, so one workspace can't spend the whole plan.
Topics now sit next to tags wherever you create or review prompts. Assign them when generating prompts from keywords or pages, see them in the prompts table, include them in exports, and filter citations, Reddit, and YouTube by topic.
You can now connect Bing Webmaster Tools next to Google Search Console in project settings. Query, page, and ranking stats sync on a schedule and feed the Search insights tab on the 360 Dashboard, so Bing traffic shows up alongside Google and AI.
Sites hosted on Kinsta can now send crawler logs through Cloudflare, using the same Worker deploy as WP Engine, Webflow, and Shopify. The crawler setup guide walks through the DNS and Worker steps.
You can now stack a persona on top of a monitor's prompts. Every prompt run is asked from that persona's perspective, so you can track how AI answers change for the exact audience you care about, like a technical buyer, a first-time shopper, or an enterprise decision maker.
Crawler visits are now verified against each AI provider's published IP ranges. Bots pretending to be ChatGPT, Claude, or Perplexity crawlers are filtered out, so your crawler analytics only count real AI crawler traffic. Visit counts may look lower than before, but the data is more accurate, and the verification applies to historic data too.
Promptwatch is now an official plugin in the ChatGPT store. Add it to ChatGPT, approve access to your organization, and your AI visibility data becomes part of the conversation: ask how your visibility changed, which sources cite you, or what your competitors are doing, and ChatGPT answers from your live Promptwatch data.
Articles published by the Content Agent now carry proper attribution. Set a default author for your project — name, title, and photo — and every publish maps it to your CMS author fields. Need a different byline on a specific article? Override the author per document before publishing.
Two Agent Chat updates: archived chats now live in a collapsed section at the bottom of the sidebar, so old conversations stay out of the way but within reach. And the agent now works with your full prompt list instead of a truncated sample, with specialist analyses returning far more detail than before.
A round of fixes across the platform to keep things running smoothly:
Connect your WordPress site to Promptwatch with our official plugin. One install brings three integrations together, with no CDN setup or code changes required:
A new Citation Trends tab shows how your citations evolve over time, so you can spot rising and fading sources instead of looking at a single snapshot.
A big batch of additions across the REST v2 API, the MCP server, and Agent Chat:
Prompt detail pages now include per-prompt charts and richer stats, and the prompts table shows each prompt's citation rate. Visibility and sentiment changes now line up with the time range preset you've selected, and you can shift-click to select a range of prompts in any list.
Put your AI search analyst on a schedule. Agent Chat can now run recurring analyses for you, like weekly briefings, competitor checks, or citation reviews, and deliver the results in-app and to Slack.
Action items are now built for teams. Assign each item to a teammate, discuss it in comments with @-mentions, and follow the full activity timeline. Assignees and mentioned teammates get notified, so nothing falls through the cracks.
Every export now offers Excel alongside CSV. Pick the XLSX format in the export modal and get a spreadsheet that opens cleanly in Excel, with no delimiter or encoding issues.
The Content Agent is now available to all plans. It plans, writes, and publishes GEO-optimized content on a schedule you control, sourcing ideas from your content gaps, prompts, and competitor data, and publishing straight to your CMS.
Query Fanouts got a lot smarter. New insights reveal the terms and themes AI models actually search for behind every answer, including a per-model comparison, and you can now see the exact fanout queries behind each individual response.
Agent Chat is now live for everyone. Ask anything about your AI visibility and get answers backed by your live Promptwatch data. The agent digs through prompts, citations, competitors, and traffic for you, and renders charts right in the conversation.
You can now filter the main dashboard by monitor, making it easy to focus on one market or product line at a time. Exports now include the monitor ID as well.
We now track brand visibility on GPT 5.6 Luna, OpenAI's newest model. Older deprecated models are automatically cleaned up from your monitors so your data stays focused on models people actually use.
A batch of additions across the REST v2 API and MCP server:
The weekly action digest email has been redesigned: top-priority items get their own cards, the layout is fully mobile-responsive, and it renders correctly in dark mode.
Accidentally deleted a project? You can now restore recently deleted projects yourself from the settings page, no support ticket needed.
Prompt visibility trends can now be exported to CSV, matching exactly what you see in the table. This joins the recently added monitor overview CSV export with synced visibility windows.
Competitor domains are now highlighted in source citations, so you can spot at a glance when AI answers cite your competitors' content instead of yours. Offsite mentions now correctly include mentions of your own brand too.
Prompt tracking keeps expanding globally: French (Belgium) is now a supported language, and Bahrain plus a range of West African countries are available as tracking locations.
When a page can't be crawled, the app now explains why, for example when a page isn't HTML or its content is too large to process, so you know whether to fix something or simply move on.
A new weekly progress email summarizes your key metrics and visibility changes per project, so your whole team sees how AI search performance is trending without opening the dashboard.
AI search now shows sponsored placements, and your competitors may already be in them. Ads Radar automatically detects ads inside AI answers and gives you a full dashboard of who's advertising, on which prompts, and where they send the traffic.
A round of fixes across the platform to keep things running smoothly:
Setting up crawler log tracking on Cloudflare is now a one-click experience. Promptwatch can deploy the crawler-log Worker to your Cloudflare account automatically, with no copy-pasting scripts or manual Worker configuration required.
You can now create read-only API keys. They work across the REST API and MCP server but block all write operations, perfect for dashboards, analysts, and AI tools that should see your data without being able to change it.
Action items that call for social engagement now include ready-to-use social post drafts, so you can go from recommendation to published post in minutes.
Shopping's My Products just got more powerful: generate tracking prompts directly from your tracked products (up to 50 at a time) and filter products by match status, so you always know how your catalog shows up in ChatGPT shopping results.
Agency plans now include unlimited projects, and pitch projects are unlimited too, so you can spin up a workspace for every prospect without worrying about quotas.
The Trends page now has a prompt search bar, so you can jump straight to the prompts you care about inside the visibility heatmap.
If you have outstanding organization invitations, they now appear right at login so joining your team's workspace is one click instead of a hunt through emails. Crawler log tracking is now also available on all agency plans.
Topics are a new way to group and analyze your tracked prompts. Assign topics to prompts and filter your analytics (visibility, sentiment, citations) by topic to see exactly how each theme performs in AI search.
Content creation now handles imagery end to end. Generate on-brand featured images with AI, using your brand assets as a reference and creative controls, or upload your own images and resize, align, and reposition them directly in the editor.
Projects can now have a display name separate from their URL slug, useful for agencies managing many client workspaces that need clean, readable names.
Visitor Analytics gained two new dashboards: Top Pages shows which pages AI-referred visitors land on most, and Top Locations breaks down where that traffic comes from.
Track pages programmatically: the new Page Tracker endpoints in the REST v2 API and MCP server let you add pages (individually or in batches with a clear added/skipped/failed breakdown) and query their citation performance.
Content gap analysis can now be re-run whenever you need fresh results, with clear requirements and progress indicators while it works. You can also filter content gaps by tag to focus on specific areas.
Crawler analytics now detects and tracks Meta AI crawlers, so you can see when Meta's models are reading your site alongside OpenAI, Anthropic, Perplexity, and Google.
A new heatmap view visualizes prompt performance at a glance: spot your strongest and weakest prompts across models and time without scanning tables.
You can now change the email address on your account directly from the dashboard settings.
Offsite mentions can now be filtered to zero in on the sources that matter, and you can generate tracking prompts directly from your existing pages, turning your content library into prompt coverage in one step.
A round of fixes across the platform to keep things running smoothly:
A pair of UI refreshes for a cleaner, more consistent experience:
Unified Actions now includes sentiment-focused recommendations that help you improve how AI models talk about your brand. You'll also get weekly AI-written insight summaries, and the action detail view has been redesigned to make each brief easier to read and act on.
Share Promptwatch and get rewarded. A new Refer page gives you a personal referral link, an earnings summary, and an activity view so you can track every referral and payout in one place.
When publishing to your CMS, you can now map Promptwatch content fields to the matching fields in your CMS, giving you precise control over exactly where titles, bodies, and metadata land.
Publish optimized content straight to your Framer site. Framer joins our growing list of CMS integrations, so you can take content from draft to live without copy-pasting.
Account owners can now opt into a weekly email that rounds up the highest-impact GEO actions to tackle next, so your team stays on top of new opportunities without logging in. Manage it anytime from settings.
Connecting Promptwatch to AI tools is now a one-click experience. Our MCP server, which lets tools like Claude and Cursor query your Promptwatch data, now supports secure OAuth sign-in, so you no longer need to copy an API key to get connected.
You can now filter tracked-page citation data by date range, and the empty state includes a date picker, making it easier to focus on the exact time period you care about.
Your browser tab now shows the name of the project you're viewing, so it's easy to tell projects apart when you have several open at once.
Discover new prompt ideas straight from Google. In the Prompt Explorer, you can now generate prompts based on related "People Also Asked" questions for your topics.
A new toggle in the response view lets you switch any AI answer between its formatted view and the raw markdown, handy for copying the exact text or inspecting how a response is structured.
The Query Fan-Out report now shows helpful guidance when there's no data yet, explaining what fan-outs are and what to do to start seeing them, so the page is useful from your very first visit.
A new My Products area in the Shopping tab lets you add the products you care about, manually or via CSV import, and track exactly how they show up across ChatGPT Search shopping results. See when your products are recommended alongside competitors and how that changes over time.
A round of fixes across the platform to keep things running smoothly:
A new toggle in the response sheet lets you highlight every mention of your brand and competitors directly inside the AI answer. Scan long responses in seconds and immediately see how each brand is positioned.
Scheduled monitor and dashboard PDFs now include the competitor heatmap, rendered in a new compact layout that fits cleanly into the report. Get a complete share-of-voice picture in every export, without opening the dashboard.
Meet the new Unified Actions tab: a kanban board where our agent surfaces the highest-impact things you should do next to grow your AI visibility. Each card opens to a single, clear AI-written brief covering what to do, why it matters, and the expected GEO impact. Action types include content gaps, untracked pages, untapped pages, Reddit social opportunities, setup gaps, and offsite mentions. New suggestions arrive in a Suggested column so you can accept or dismiss each one, then move them through To Do, In Progress, and Done as you work through them.
AI Overview responses now render product thumbnails, star ratings, and rich snippet details inline, giving you a much closer view of exactly what searchers see in Google's AI Overviews.
GET /responses and GET /responses/:id now return per-brand mention details plus citation sourceType and contentType, so BigQuery and other data pipelines can pull everything in one paginated call. A new GET /api/v2/sitemap/urls endpoint also exposes errored, redirected, and in-progress URLs, and the Postman collection has been refreshed.
The Knowledge Base now opens with step-by-step guidance that explains what it does and how AI uses it for content generation, so the first time you visit you know exactly what to add and why.
Brand views across the dashboard now default to your tracked brands and direct competitors, instead of showing every brand that's ever been mentioned. Cleaner numbers, less noise, and you can still switch to the full list anytime.
Sponsored product cards in AI Overview responses now appear at the bottom of the response instead of mixed into the editorial content, so it's easier to read the organic answer first.
A new Shopping tab surfaces every product card, price, merchant, and review extracted from ChatGPT Search responses for your tracked prompts. See exactly which products AI is recommending alongside your brand and competitors, track how that mix shifts over time, and spot category trends you'd otherwise miss. Available on Professional plans and above.
The usage modal now shows a per-project breakdown so you can see exactly where your prompts, monitors, and credits are being spent across projects in your organization.
You can now select a language each time you generate content, in addition to a default project language. Generate articles, briefs, and other content in any of our supported languages without changing project settings.
The brand prompt list in the overview now only includes direct competitors instead of every brand mention. This keeps the view focused on brands that actually matter for your competitive analysis.
A new search bar on the Reddit page lets you search across every tracked Reddit post, username, and subreddit in your project, making it much faster to find specific conversations and creators mentioning your brand.
Matched brand aliases are now visually highlighted within AI responses, so you can immediately see exactly which alternative names and spellings the AI used when mentioning your brand or competitors.
You can now choose a default tone of voice for AI-generated content right when you create a project or during onboarding. The selected tone is automatically applied as a default for content generation, so every article matches your brand voice from day one.
Trying to delete a brand tag that's still attached to a project now shows a clear error message explaining why, instead of silently failing or producing a confusing error.
Monitor spark charts no longer look janky when you only have a small number of data points. New monitors now render clean, readable charts from the very first datapoint.
The offsite mentions table now shows a dynamic skeleton while loading and displays a clearer empty state when no mentions are found, eliminating layout shifts and confusion during page loads.
Select multiple prompts in the prompts table and update their intent and type in a single action. Reorganizing large prompt libraries is now far quicker.
Scheduled PDF report exports are now available to all customers (no longer behind a feature flag). Set up recurring monitor and dashboard reports delivered straight to your inbox, and see when each schedule will next run from the reports table.
Bookmark important pages from the citation and grouped citation tables to quickly jump back to your most-watched URLs. The Tracked Pages experience has also been refreshed with clearer terminology throughout the app.
Filter chips now use the same model labels as the dropdowns, so the model you selected is shown the same way everywhere across the app.
The Top Cited Domains filter now applies correctly, returning the right results when narrowing down citation data.
A new stacked chart visualization makes it easier to see how mentions are distributed across brands and competitors over time at a glance.
The content agent modal has been refined with a cleaner layout and better controls, making it easier to configure and run content generations.
Tag prompt counts are now correct and consistent with the prompts table, so the numbers you see on each tag actually match the prompts inside.
Removed redundant +/- signs from chart comparison values so deltas are easier to read at a glance.
Projects can now track multiple sitemaps instead of just one. Add all your sitemap URLs and get a unified view of AI crawler activity across your entire site.
When adding a competitor, entering the URL first now automatically fills in the competitor name, saving you a step during setup.
The Query Fan-Out report has been redesigned with a clearer layout and better charts, making it easier to understand how AI models expand your queries into sub-queries.
Citation collection now runs faster and handles queue retries more reliably, helping your citation insights update sooner and with fewer stalled scrape jobs.
Project onboarding now skips repeated website scrape attempts after multiple failures, so you can continue setup without getting blocked by a failing site fetch.
Connecting Google Search Console is now clearer and easier to complete, reducing setup friction when linking your project data.
Track your brand visibility in Cambodia. We've added Cambodia as a monitor country option along with Khmer language support for prompts.
ChatGPT sometimes returns citation markers as plain ASCII text instead of formatted tokens. Citations now render correctly regardless of the format ChatGPT uses, so you always see proper citation badges instead of raw text like "citeturn0search0".
AI-generated content now streams more smoothly in the editor, with tokens appearing at a natural pace instead of in choppy bursts.
The content editor now disables the submit button when the prompt is empty, limits AI text selection to 2,500 characters to prevent oversized requests, and the prompts table shows when each prompt was created.
Track your brand visibility in even more markets. We've added Lithuanian, Latvian, Ukrainian, and Chinese (Simplified and Traditional) language options for prompts and monitors.
You can now add brand aliases right when creating a new project or during onboarding. No need to visit Brand Book separately: set up alternative names and spellings from the start.
Filter your offsite mentions by type and source to zero in on the third-party citations that matter most. Quickly find mentions from specific content types or platforms.
Track your brand visibility in Kazakhstan with the new Kazakh language option for prompts and monitors.
The Sitemap page is now accessible directly from the main app sidebar, making it quicker to check AI crawler activity on your sitemap.
Generate monitor and dashboard PDF reports with a unified export flow, preview report settings before exporting, and optionally receive an email when the report is ready to download and share.
The content optimization modal now features a sidebar layout that makes it easier to navigate between sections. Required settings are shown first, and you can save and go back between steps without losing your progress.
Generate detailed PDF reports for individual monitors. Share performance data, visibility scores, and competitor comparisons with stakeholders in a clean, printable format.
Our full list of known AI crawler user agents is now published in the docs and in-app. Copy individual user agents or the entire list to configure your own crawler log filtering.
Connect your AI coding tools directly to your Promptwatch data. Our new MCP (Model Context Protocol) server lets you query visibility scores, citations, prompts, and more from tools like Cursor and Claude Desktop, right in your workflow.
Added new Google-Agent user agents to crawler detection. Your crawler logs now identify and track the latest Google AI crawlers accessing your content.
AI Mode is no longer restricted to specific countries. You can now select AI Mode as a search type for monitors in any supported country.
Generate and optimize content programmatically through the REST API v2. Use the same AI-powered content creation tools available in the dashboard directly from your own applications and workflows.
Track how often your brand is mentioned in AI responses over time. A new mentions chart on the visibility dashboard shows trends in mention frequency, making it easier to spot growth or drops at a glance.
See where your brand ranks in AI responses. A new chart shows your average position and rank across prompts over time, helping you understand whether AI models are listing you first or further down.
Google Search Console connections now work at the project level instead of the organization level. Connect multiple GSC properties to different projects and import topics directly from Search Console into the Prompt Explorer.
A round of fixes to keep things running smoothly: visibility scores now filter correctly when tags are applied, brand merging no longer times out for brands with many similar names, deleted projects no longer cause redirect loops, and AI responses are sorted by newest first by default.
Two new dedicated reports are now available. The Query Fan-Out report shows how AI models expand your queries into sub-queries, and the YouTube Citation report lets you see which videos are being cited, how often, and by which AI models.
Dive deeper into how third-party sites are cited alongside your brand. New stats and breakdowns show which offsite sources appear most frequently in AI responses, helping you understand the competitive citation landscape.
Understand how AI models position your brand relative to competitors. A new Brand Positioning view shows where your brand stands in AI-generated recommendations and comparisons.
You can now assign tags to prompts right when you create them, including during batch uploads. A new review step also lets you double-check prompts before they're added.
Turn content gap insights into action. You can now create new content or optimize existing pages directly from Content Gap recommendations, with full control over generation settings.
Download your offsite citation data as a CSV file. Analyze which third-party sources are cited alongside your brand in spreadsheets or share with your team.
Filter out internal traffic from your visitor analytics by adding known IP addresses to an ignore list. Get cleaner data by excluding your own team's visits.
Connect your Amazon CloudFront CDN logs to track AI crawler activity. See when AI systems access your content delivered through CloudFront.
The sitemap page now shows a clearer breakdown of crawler activity with toast notifications for key actions. Easier to understand how AI bots are interacting with your sitemap.
API keys are now displayed clearly with a copy button for quick access. No more manual selecting and copying, just click to copy your key to the clipboard.
A batch of fixes to improve reliability across the platform: ghost citations from deleted prompts no longer appear in results, brand visibility charts correctly distinguish brands with similar names and subdomains, project viewers no longer receive unrelated organization notifications, responses table text no longer overflows, time range filters work correctly, content gap metadata no longer leaks into text reports, and several crashes and UI glitches have been resolved.
Three new CDN integrations for crawler log tracking. Connect your Netlify, Vercel, or Google Cloud Platform logs to see when AI crawlers access your content, no matter where it's hosted.
Sitemap crawls now surface more detailed information about each page, giving you deeper insight into how your sitemap is being processed and indexed.
Tables across the app, including Reddit citations, YouTube mentions, and AI responses, now have improved column sizing and no longer cause horizontal scrollbars. Data is easier to scan at a glance.
A new comparison chart lets you see how your brand visibility has changed between two time periods. Easily spot trends and measure the impact of your optimization efforts over time.
Quickly copy any prompt's text to your clipboard with a single click. Handy for sharing prompts with teammates or testing them in AI tools directly.
AI responses that reference news articles are now tracked as a distinct citation type. See when ChatGPT pulls in news sources alongside your brand.
Prompts generated from your website now include a review step before they're added. Preview and edit auto-generated prompts to ensure they match your tracking goals.
Visibility charts no longer overflow their container on smaller screens. Charts now resize properly for a better experience on any device.
A new calendar view gives you a visual overview of your content publishing schedule. See when content was created or updated and plan your optimization efforts over time.
Key metrics across the platform now have more descriptive labels, making it easier to understand what each number represents at a glance.
A dedicated report for Reddit citations is now available. See which Reddit posts and subreddits are being cited in AI responses about your brand, with detailed breakdowns and trends.
Track your brand visibility in even more regions. We've expanded our country selection with additional countries so you can monitor AI responses in more markets around the world.
Download your Content Gap analysis results as a CSV file. Easily share coverage scores and content insights with your team or analyze them in spreadsheets.
Your dashboard now includes a competitor heatmap showing how your brand compares to competitors across prompts at a glance. Quickly spot where you're winning and where there's room to improve.
Never lose your work when creating prompts. You'll now see a confirmation dialog if you try to close a prompt creation form that has unsaved changes.
Newly created prompts are now prioritized in the analysis queue. Get your first results faster when you add new prompts to a monitor.
We've improved how we crawl and process your sitemap. Pages are discovered more reliably, and sitemap data is cleaned up for better accuracy in page tracking.
We've restructured the sidebar navigation to make features easier to find. Social mentions (Reddit and YouTube) are now grouped together, and settings are more logically organized.
Share your AI visibility insights with stakeholders by exporting your dashboard as a PDF report. Perfect for presentations and executive summaries.
The offsite citations page has been reorganized for better readability. Citation data is now easier to scan and compare at a glance.
Custom HTTP endpoints for crawler logs now support sending multiple events in a single request. Reduces overhead and makes large-scale log ingestion more efficient.
A new chart breaks down your citations by type: see how many are direct mentions, links, or other citation formats across AI responses.
Visitor analytics now tracks traffic from Mistral AI (chat.mistral.ai). See how many visitors are coming to your site from Mistral's AI assistant.
Track your brand visibility in even more languages. We've added support for additional languages to help you monitor AI responses in more markets around the world.
Filter crawler logs with more precision. You can now choose between exact path matching or partial matching to find the specific pages AI crawlers are visiting.
Visitor analytics now combines traffic from copilot.com and copilot.microsoft.com into a single Copilot source. Get a complete picture of your Microsoft Copilot referral traffic.
Understand your metrics better with new in-app help. Click on any key metric like Visibility Score, Brand Sentiment, or Content Coverage to see an interactive explanation of how it works and what it means.
Track AI Overviews from Google based on your target location. See how AI search results appear in different regions and languages to better understand your local visibility.
Quickly find relevant citations using your prompt tags. The citation page now includes a tag filter so you can focus on the prompts that matter most.
Quickly select or deselect all AI models in the filter dropdown. Save time when you want to view results across all models or start fresh with your selection.
Add alternative names and spellings for your brand in Brand Book. Whether your brand is known as "Acme Corp", "ACME", or "Acme Incorporated", we'll now detect all variations in AI responses.
Content optimization now uses more sources to create better content. We include Google News for timely information, your site's internal links for better navigation, query fan-outs for comprehensive coverage, and brand-specific context for authentic voice.
Instantly spot brand mentions in AI responses. Your brand and competitors are now visually highlighted, making it easier to see how each is discussed.
View images from ChatGPT responses in an interactive carousel. Browse through all images included in AI responses with easy navigation.
Your filter selections now persist as you navigate between crawler log pages. No need to re-apply filters when switching between different views.
See when other brands are mentioned alongside yours in AI citations. Understand your competitive landscape and discover which brands appear in the same context as yours.
See which pages on your site AI crawlers visit most frequently. A new analytics view shows your top crawled pages with detailed visit statistics and CSV export.
Easily set up visitor analytics using our Google Tag Manager template. A simpler way to track AI-powered visits to your website.
Finding prompts is now faster and more accurate. Search works better with partial matches and multi-word queries.
Crawler logs now display crawlers sorted by visit frequency, and long page paths are properly truncated for better readability.
You can now choose your preferred language for prompts during the onboarding process. Get started tracking in your target market right away.
Comprehensive API documentation is now available. Explore endpoints, authentication, and examples to integrate Promptwatch data into your workflows.
CSV exports now include additional data fields. Get more detailed insights when analyzing your AI visibility data in spreadsheets.
Discover what's missing from your content. Our new Content Gap feature analyzes your pages against AI prompts to show where your content falls short. See a coverage score for each page and get actionable insights to improve your AI visibility.
We've combined intent and type filters into a single, easier-to-use control. Filter your prompts faster with fewer clicks.
Monitor pages now feature improved charts including a prominent competitor comparison chart and sentiment tracking. See your brand's performance at a glance.
You can now sort your prompts alphabetically by name. Quickly find and organize prompts in large lists.
Click on top citations in your dashboard to instantly see more details. Access citation information without leaving the page.
Monitor how specific pages on your website perform in AI search. See which pages get cited most and optimize your content strategy accordingly.
Query fan-outs now display which AI model generated them. See at a glance whether ChatGPT, Claude, or another model expanded your query.
When you create a new project, your homepage is now automatically added to tracked pages. One less step to start monitoring your AI visibility.
Click on any tracked page to see detailed citation information in a side panel. Quickly understand how each URL is being cited across AI responses.
The tracked pages table now supports sorting by different columns and displays sparkcharts showing citation trends over time.
Quickly access content optimization suggestions directly from your prompts. A new button lets you start improving your visibility with one click.
Crawler logs now load significantly faster thanks to optimized database queries. Large datasets are no longer slow to display.
Track your visibility across the latest Claude AI models. Stay ahead by monitoring how newer versions respond to your brand.
You can now move prompts from one monitor to another. Reorganize your tracking setup without having to recreate prompts from scratch.
CSV exports now include brand mentions and match your table columns exactly. We've also added export support for query fan-outs, making it easier to analyze your data in spreadsheets.
Track your brand visibility in Thai and Vietnamese AI responses. Expand your monitoring to these growing markets.
See helpful stats about your content at a glance, including word count, reading time, and link count. A new visual gauge helps you understand your content's optimization level.
Fixed an issue where adding links to selected text in the content editor wasn't working properly.
You can now browse and explore all your prompts in one place. Filter by category, search across monitors, and quickly find the prompts that matter most to your visibility tracking.
Enterprise teams can now use their company's identity provider to sign in. SSO makes it easier to manage team access and maintain security compliance.
Try our new free tool that shows you how ChatGPT expands your prompts into multiple search queries. Understand how AI interprets your questions and optimize your content accordingly.
Export your YouTube and Reddit citation data to CSV. Perfect for deeper analysis in spreadsheets or sharing insights with your team.
Getting started with Promptwatch is now easier than ever. We've redesigned the onboarding flow with clearer steps, better guidance, and a smoother setup process.
Track how positively or negatively AI models talk about your brand over time. New sentiment charts help you monitor your brand perception in AI responses.
See when AI models cite YouTube videos in their responses. Track your video content's visibility across ChatGPT, Claude, and other AI search engines.
Monitor how often Reddit posts mentioning your brand appear in AI responses. Understand your Reddit presence in the AI search landscape.
Explore how ChatGPT recommends products and influences shopping decisions. See which brands and products get mentioned in shopping-related queries.
See all your citations organized by domain. Quickly understand which websites are being cited alongside your content and identify new opportunities.
Get a free analysis of how well your website is optimized for AI search engines. See what AI models can read on your pages and get actionable recommendations.
Preview how your website appears when shared on social media and in AI responses. Check your meta tags, images, and descriptions in one place.
Our AI-powered content optimization feature is now available to all users. Get suggestions to improve your content's visibility in AI search results.
Learn how to optimize for AI search with our new Academy. Guides, tutorials, and best practices to help you improve your AI visibility.
Download your AI crawler activity data for deeper analysis. Export logs to CSV and integrate with your existing analytics workflows.
We're excited to announce our $1.4M seed funding round led by Arches Capital. This investment will accelerate our mission to help businesses optimize their visibility in AI search engines.
Add tags to your prompts to keep them organized. Filter and search by tags to quickly find the prompts you need.
Filter your data by custom date ranges. Select specific time periods to analyze trends and compare performance across different timeframes.
Export your optimized content in the format you need. Download as Markdown for your CMS or HTML for direct publishing.
Create and edit content directly in Promptwatch with our new rich text editor. Format your content, add headings, and prepare it for publishing - all in one place.
Track how your competitors appear in AI responses alongside your brand. See who's getting mentioned and compare your visibility against the competition.
Connect your server logs to see exactly when AI crawlers visit your website. Understand how often ChatGPT, Claude, and other AI systems are indexing your content.
See your website traffic from AI sources in real-time. Track visits from ChatGPT, Perplexity, and other AI-powered browsers with our new analytics dashboard.
A new step-by-step checklist guides you through setting up your account. Never miss an important configuration step again.
Connect your Fastly CDN logs to Promptwatch for real-time AI crawler tracking. See when AI systems access your cached content.
Export your prompts, responses, and citations to CSV. Analyze your data in spreadsheets or import it into other tools.
Get a free report on how your brand appears in AI search results. Enter your brand name and see how ChatGPT and other AI models talk about you.
Access your Promptwatch data programmatically with our new API. Create and manage API keys to integrate AI visibility data into your own tools.
Quickly create new monitors by copying existing ones. Save time when setting up similar tracking configurations.
Check if AI crawlers can access and read your pages. Our free tool shows you exactly what AI systems see when they visit your URLs.
Create an llms.txt file for your website with our free tool. Help AI models understand and properly cite your content.
Collaborate with your team on AI visibility. Invite colleagues to your Promptwatch workspace and work together on optimizing your AI search presence.
Discover questions AI models struggle to answer about your industry. Find content opportunities where you can become the go-to source.
Quickly filter your citation data to see which responses mention specific competitors. Understand the competitive landscape in AI search.
Want a personalized walkthrough? Schedule a demo with our team to see how Promptwatch can help your specific use case.
See detailed analytics on where AI models cite your content. Track citation trends, compare sources, and understand your visibility patterns.
Today marks the official launch of Promptwatch! Our platform helps businesses track and improve their visibility across AI search engines including ChatGPT, Claude, Perplexity, and Gemini.
Join thousands of companies using Promptwatch to monitor and improve their presence in AI search results.
