Impact now shows which leads and sales were traced to an AI-referred visit, and how each one was traced. A new funnel sets the AI-referred visits in a window beside the leads and revenue traced to them, by engine and by landing page. Set it up in Traffic → Setup.
The funnel counts each stage on its own: AI-referred visits, the leads traced out of every lead, and the revenue traced out of all revenue, with what wasn’t traced shown beside it. No conversion rate is drawn between them.
Each traced receipt says in words how it was traced, dated in UTC: by the visit’s own id that it carried, by the same visitor that day, through the customer’s first lead, by a checkout’s client_reference_id, or, for a renewal, through the customer’s first checkout.
Two opt-in pixel helpers carry the visit’s id to where your systems can send it back: data-decorate="stripe" adds it to Stripe Payment Links as their client_reference_id, and data-trovance fills a hidden field in your lead forms. Neither sends anything itself or stores anything in your visitors’ browsers.
Every revenue figure counts one source: Stripe while a Stripe webhook is connected with its currency saved, otherwise your server. The other source’s revenue is shown beside the funnel, never added, so a sale reported both ways is counted once. Refunds are not subtracted.
The funnel on Impact’s Overview · example dataThe setup guide in Traffic → Setup · example data
Recommendations now weigh your search, analytics, AI answers, competitor moves, and connected business data together, and explain why each action is worth taking. After a change goes live, Trovance follows up at 7, 28, and 56 days and turns comparable results, including the ones that didn’t work, into lessons for later recommendations and briefs.
Pages you already have count too: Search Console queries and site engagement can point to a specific improvement on an existing page.
Each recommended action shows why it was chosen, the intended change, and how it will be checked. A source that is missing or out of date is named, not counted as zero.
Results read positive, negative, mixed, or inconclusive. They record what changed, not proof that one change caused it, and a lesson needs at least 28 days of comparable observations.
In Results, see what informed each move, withdraw a publication’s lessons, or leave feedback for the next cycle. The assistant answers from the same evidence.
The release film · example brand
What informed the next move, in Results · example data
The platform now matches the rest of Trovance. Primary actions are lime, light mode is neutral, and Home sits on a quiet pixel grid. Navigation, layouts, and workflows are unchanged.
Main buttons, including Create, Ask, and Create draft, use lime with dark text in light and dark mode. Teal stays for links, charts, and status.
Light mode drops its green tint for neutral whites and grays.
Home’s grid backdrop drifts slowly, freezes while you type, and stays still when your device asks for reduced motion.
The first-run introduction and the Create dialog use the website’s forest, lake, and river imagery, drawn in teal dots.
New drafts default to your brand’s voice and close by connecting the topic to your brand, with one next step. Recommendations follow the pillars you set on the Authority Map.
Drafts are written in Brand voice, from Brand Core, unless you choose an author. Author voices and your personal voice stay within the brand you’re working in.
Each new article closes after its FAQs with a short section on how your brand relates to the reader’s decision, ending with one call to action.
That link goes only to your homepage or a page already recorded from your site. Without a website on file, the section ends with a plain invitation instead.
Editing pillars refreshes recommendations: questions move under your current pillars, and undecided suggestions tied to a removed or renamed pillar are retired. Pinned questions and decided work stay as they are.
Start with Trovance from iMessage on an iPhone, iPad, or Mac. New visitors get a first look at how AI answers treat their company, plus one complete draft. Customers can connect a workspace and keep working by text.
Text Trovance on the homepage opens Messages with your note written. Send your website and one priority, and Trovance runs three buyer questions through AI engines, shows how often you appeared in the sampled answers, and suggests where to start.
Reply for one complete draft, read in Messages. Editing, publishing, and more drafts begin with a 14-day trial through Stripe checkout. Sending a message never bills you.
Already a customer? Choose Connect iMessage under Settings → Notifications to link your phone to your workspace, with your current permissions. Ask questions, draft and revise, review recommendations, import a file into Brain, or publish to a connected WordPress or Framer site.
Nothing changes without your exact code reply, and codes expire. Trovance replies only in conversations you start, and STOP pauses messages without cancelling a subscription.
Connect iMessage from Settings → NotificationsLink your phone with a one-time code · example brand
Import docs in Brain now reads PDFs, Office and OpenDocument files, spreadsheets, web pages, ebooks, and images. Each becomes a readable copy you review, and nothing joins Brain until you add it.
Supported: PDF, DOCX, PPTX, XLSX, ODT, ODP, ODS, RTF, CSV, TSV, plain text, Markdown, HTML, EPUB, PNG, JPG, and WebP. Up to 5 files at a time, 20 MB each, and 7 MB per image.
Review the readable copy, download it as Markdown, or open the original, then choose Add to Brain.
Scanned pages and images are read by AI only when you allow it for that upload.
Your assistant and writer draw on relevant passages. Removing a document leaves past drafts as they are.
Import documents, and decide how scans are read · example document
Recommendations you can draft from now show related work: up to two existing pieces that may cover the same idea, so you can check them before writing something new.
Related work can be a draft, an accepted brief, approved or scheduled work, or a page on your site, each with its status, a date, and a short excerpt.
Open the draft, the work, or the published article straight from the review.
The check covers your saved pages and active pieces, and says how many it read. It isn’t a scan of your entire website.
When Mailchimp or Klaviyo is connected and one recent email clearly earned a higher click rate than the rest, Trovance suggests turning its topic into an article. The suggestion appears in Home’s next actions, and you review the evidence and shape a reader brief before anything is drafted.
The evidence is dated: click rates for your top campaign messages over the synced period, counted as unique clickers per delivered email. It points to reader interest; it isn’t traffic or revenue attribution.
Before drafting, set a reader-facing title, who the piece is for, and its direction, then choose your Brand Core voice or an author’s voice.
Campaign figures help choose the topic but stay out of the draft. The writer gets only the brief you approve.
A suggestion needs a sync from the last 48 hours, at least three campaign messages with 100 or more deliveries each, and a clear lead in click rate. Until then, Home says what it’s waiting for.
The evidence, by campaign click rate · example dataThe reader brief you approve · example data
Bring more of what your team knows into Trovance. New connections for Notion, Google Drive, and Attio add source material and sales context to the work you are already doing.
Choose a Notion page, Google Doc, or Slides presentation, review its text, and explicitly import a copy into your Brain.
Imported material keeps a link to its source. Copies stay in your Brain after disconnecting; changes to the original are not automatically synced.
Attio brings in aggregate counts by stage for deals created in the last 30 days, so recent sales activity can inform planning. Your Attio workspace needs the Deals object enabled.
Find the new connections in Settings → Connections.
Choose a Notion page · example source materialReview a Google Drive import before adding it to your Brain · example source material
HubSpot, Klaviyo, and Mailchimp join Connections. Recent sales and email performance can now inform content planning, while a new setup flow makes connecting your tools easier to follow.
HubSpot supplies recent deal counts by pipeline stage. Klaviyo and Mailchimp supply campaign topics and aggregate delivery and click performance.
Connected business data refreshes daily. These connections provide planning context; they do not send campaigns or publish content.
Short setup steps explain what each connection needs, show what access it uses, and retain entered details when you go back.
Squarespace gets a guided handoff: prepare the article in Trovance, publish it in Squarespace, then record its live URL.
HubSpot connection setupA guided Squarespace publishing handoff · example article
New customers can read one complete first draft before providing a card. Your 14-day trial starts when you activate the draft through checkout, giving you time to see the work before deciding to continue.
Activation unlocks editing, Fix with AI, additional drafts, and publishing after checkout is confirmed.
Choose the paid plan during checkout. Inviting a teammate unlocks Growth features for the trial without changing that choice or extending the trial.
Existing customers keep their current terms.
A complete first draft, with trial activation in your hands · example article
Your first opportunity, draft, live article, and results now stay connected. An optional setup guide helps you move through the first week, and each published piece has a workspace for deciding what comes next.
Pick up the active draft or published piece from Home, tracking setup, or your weekly update.
Carry the current article into publishing setup and return to the same piece when the connection is ready.
See publication verification, page traffic, AI referral visits, conversions, and comparable answer evidence together.
Review the next check, keep observing, or use the evidence to shape the next brief. Waiting for data, missing tracking, and unavailable readings are shown separately.
A published piece, its results, and the next decision · example data
Home leads with your question, the work in progress, and the next useful action. The surrounding workspace has been refined to make finding, planning, and measuring that work easier.
Add a file, link, or pasted text from the composer without losing your unfinished prompt. Actions puts existing workflows and suggested questions within reach.
Trovance landscapes change with the day, with compact readings and a direct route back to your conversation or active piece.
Library adds search, filters, pagination, and record previews. Traffic separates reports from tracking setup and adds page details.
Calendar adds week and agenda views alongside the month view. Planning a date does not automatically publish the article.
Home keeps your current work close · example workspacePlan content across month, week, and agenda views · example schedule
Brand Core is now a working brief for your team and Trovance. Positioning, audiences, offerings, voice, and supporting resources live together, with your saved decisions guiding the work across the platform.
Start with a brand overview, add your logo, and export the brief for your team.
Keep detailed audience profiles, offerings, practical voice guidance, and Brain-backed resources in one place.
Suggested updates stay separate from your saved foundation. Compare current and proposed wording, then apply only the changes you choose.
The writer, assistant, and recommendations draw from the same saved foundation.
Your shared brand foundation · example brandReview suggestions before changing your foundation · example brand
Trovance is operable by agents. A discovery document at app.trovance.ai/.well-known/mcp.json lists every tool the MCP endpoint serves, derived from the same registry the endpoint enforces.
MCP endpoint at /api/mcp/mcp, authorized by a bearer API key minted under Settings → Agent access
Twenty-one tools: ten reads across visibility, answers, the queue, traffic, the authority map and the Brain; one brief suggestion; ten operations, each gated by the key’s scope and revocable at any time
The score became an index with a methodology you can read. How each check is scored is public at app.trovance.ai/methodology/agentic-web-health, so a number is never the whole story.
The methodology page: every check, and what it reads on your site
On-page fixes as recommendations: page-fix cards, end to end (Aug 12)
A lead is definable, not just a purchase: define the conversions your pixel counts (Aug 17)
Coverage caught up with the market. Weekly sweeps grew from two engines to five: ChatGPT and Perplexity on Starter, with Google AI Overviews, Gemini and Claude added on Growth and Agency.
Google AI Overviews, read through SERP data (Jul 20)
Gemini enabled on the weekly sweep, with a per-plan engine cap (Jul 21)
Claude memory lane on every plan: how you’re represented across remembered context
Share of answer per engine, so one engine’s blind spot can’t hide
Proof moved on-site. A cookieless, first-party pixel counts human and AI visits to your pages and tells agents from humans without fingerprinting anyone.
Cookieless, first-party, one script tag (Jul 12)
AI crawlers named by the middleware snippet, so agent visits and human visits are told apart
Live visitor toasts and the activity feed: watch it happen, as it happens (Jul 14)
Agent journeys: crawl events become bot sessions on Traffic (Jul 18)
Live visitor traffic · Example dataAI crawler journeys · Example data
Point the engine at a goal and it keeps the queue warm. Missions landed on July 3; on July 12 the MCP server let outside agents read the engine and make one safe write, behind an API key.
Standing missions: 1 on Starter, 5 on Growth and Agency
Research-first agent drafting, structured and sourced
The MCP server: seven reads and one safe write, keyed and revocable (Jul 12)
Every action leaves a receipt a human can audit: the activity ledger (Jul 14)
Standing missions · Example dataAgent activity history · Example data
The market becomes a map. The content graph was redrawn as an authority map: your tracked questions charted as clusters, with the ground you hold, the ground rivals hold, and the open territory between.
The authority map: clusters, routes and flags across your market
Pillars are subject territories, four to eight per brand, never funnel stages
Your company, your market and your rivals assessed on every cluster, from observed answers
From findings to a production line. Observed answers become a ranked queue of content worth making, drafted in your voice, approved by a human, published with receipts.
The ranked queue: impact-ordered, evidence-backed, and “no action” is a valid result
Author voices, trained on your edits
Governed publishing to WordPress and signed webhooks, receipts on every push (connectors, Jul 20)
Verify: the next weekly sweep re-observes the question and compares the answer against the record
Ranked content queue · Example dataContent recommendation · Example data
Point your iPhone’s camera here. Messages opens with the note written.
Or, in iMessage, send this note to +1 (839) 205-9508:
Hi Trovance, show me how AI sees my company.
You choose to send it. Trovance replies, and sends the updates you ask for, over iMessage through our messaging provider. Reply STOP anytime. Tapping or sending never bills you.