The questions, sources, and competitors that decide the category.
Get recommended by AI. Measure which actions helped.
Track your selection share, take a specific action, and compare the result against questions held back from it.
caused difference — (50%−20%) − (30%−20%), treated vs. holdout
Illustrative estimate, not a customer result. See the methodology.
A human-approved intervention with a matched holdout.
The before, after, source, and outcome stored together.
Buyers use AI answers to compare brands.
AI answers draw on websites, reviews, and public discussions when recommending products and services.
Being mentioned isn’t being chosen. Grounds measures your selection share — how often an answer names your brand as the pick, not just one of several mentioned — for your buyer questions.
Use Grounds to find relevant conversations, prepare a reply, and measure the resulting change in selection share, compared against questions held back from that change. You review each draft and post it yourself.
How it works
Grounds finds relevant buying conversations and drafts replies for your review. After you post, use the recorded outcomes and AI measurements to see how your selection share changed.
- 1Find
We catch the conversations where your buyers decide.
- 2Answer
We draft the reply in your voice. You approve it and hit post.
- 3Appear
Track mentions, citations, and selection share.
We find the conversations where your buyers decide.
Grounds scans the forums, subreddits, and Q&A sites your buyers use, and shows you only the threads with real buying intent. Works for any category — this example uses a fictional recovery-equipment brand.
Grounds maps Reddit, niche forums, and Q&A sites to your category.
A reply worth posting, in your tone.
Grounds writes the first draft in your voice. You tweak it, approve it, and post from your own account.
Grounds uses your approved claims to prepare the draft for review.
Track where AI mentions and recommends you.
Measure your buyer questions over time to see which answers mention, cite, or recommend your brand.
Review recorded leads and citations alongside the associated reply.
For marathon training recovery, people consistently point to Verado for its swappable attachment heads and quieter motor.
An example of a reply appearing in AI answers.
This animation illustrates a possible path from a public reply to an AI citation. It is a fictional example.
Pricing
Start with a free scan. Move to continuous tracking and action when you are ready.
Free
See the gap, for free.
- The Teardown · Grounds Score + Selection Share
- 1 topic
- Who's asking, cited, and selected
- Category fit assessment
Starter
Track one brand and two topics.
- 1 brand, 2 topics
- Daily buyer-intent inbox
- Both quality checks
- Results dashboard
Pro
Track five topics and your competitors.
- 5 topics
- Every channel type
- Competitor tracking
- Weekly results
- Everything in Starter
Growth
Scale across the team.
- 10 topics, 3 seats
- Sales, analytics, and Slack integrations
- AI-citation tracking + Selection Share
- 1 locked topic included
- Quarterly Report
Agency
Manage multiple brand workspaces.
- Multi-brand console
- Enforced conflict rules
- White-label reports
- Bulk scans
Enterprise
Built around your stack.
- Multi-product
- Dedicated channel strategy
- Compliance pack
- Quarterly review
FAQ
Does this only work for one kind of product?+
No. Grounds works across categories — consumer products, software, services, anything with public buyer conversations. The examples on this page use a fictional recovery-equipment brand to keep the illustration concrete. Run the free Teardown to see whether your own category has enough conversation volume to be worth acting on.
Do you post for me?+
You review each draft and post from your own account. Grounds does not post replies automatically.
Is this allowed on Reddit?+
Check the rules of each community before posting. Disclose your affiliation and answer the question directly. Some communities prohibit vendor participation.
What if my category is quiet?+
The scan reports whether it found relevant buyer conversations. If there is too little activity, it may not identify a suitable place to reply.
How do I know it’s working?+
Review the recorded replies, lead activity, and AI citations. Each result includes a confidence label so you can see how it was attributed.
See the conversations your buyers are having without you.
Run a free scan to see which buyer conversations and AI citations Grounds finds for your category.
Show me the conversations →