The list is stale. The emails bounce. And when they check you out, the AI recommends someone else.
Scout finds buyers who actually exist, writes outreach worth replying to, and shows you what the AI engines say about you when the prospect goes looking. Both halves of the funnel, one place.
No credit card required · Free web-sourced leads on day one
Heads of Growth at Series B fintechs in the US
Interactive Demo Scenarios
Workflow: Find buyers - Query: "Heads of Growth at Series B fintechs in the US"
Executed steps: Searching live sources, Enriching profiles, Verifying emails, Scoring against your ICP
- Priya Kapoor (VP Growth · Nimbus Cloud · 240 staff) - Status: Verified, Fit score: 94%
- Daniel Osei (Head of RevOps · Lattice Pay · impressions up 3x) - Status: Verified, Fit score: 88%
- Wei Chen (Director Sales · Forma Health · 95 staff) - Status: Risky - held back, Fit score: 81%
Outcome summary: 1 address looked risky, so it was held back instead of burning your domain.
Workflow: See what AI says - Query: "What does AI tell buyers when they ask about us?"
Executed steps: Asking Gemini, Asking Groq, Parsing brand mentions, Computing confidence
- Gemini: You appear in 7 of 12 answers (Rate: 58%) - 95% CI 32-81%, n=12 · usually listed after Apollo
- Groq: You appear in 2 of 11 answers (Rate: 18%) - 95% CI 5-48%, n=11 · Clay owns this answer
- Both: Change vs last week (Rate: No call) - Intervals overlap - not enough data to claim a move
Outcome summary: Engines disagree. A blended headline number would describe no engine that exists.
Workflow: Catch the moment - Query: "D2C brands that just raised and are hiring supply chain"
Executed steps: Watching funding feeds, Matching to your ICP, Finding decision makers, Drafting first touch
- Aarav Mehta (COO · Loomly Goods · raised $18M, 6 days ago) - Status: Funding signal, Fit score: 96%
- Sofia Ramos (Head of Ops · Brightbar · 4 supply chain roles open) - Status: Hiring signal, Fit score: 90%
- Tomas Lind (Founder · Nordkit · new 3PL partnership) - Status: Verified, Fit score: 83%
Outcome summary: Reached while the budget is still being decided, not three months after.
Zero Stale Exports
Queries the open web and validates live MX records and domains in real time, not year-old static dumps.
Defensible AEO Stats
Every AI citation and visibility score includes Wilson score confidence intervals and verbatim model answers.
Domain Protection
Autonomous ramp guardrails, bounce sentinels, and mailbox health checks stop sends before your sender domain burns.
First-Class AI Agents
AI agents and automated workflows access the exact same search, enrichment, and verification endpoints as human SDRs.
Ninety seconds, end to end
What Scout does to a list, what it refuses to send, and what the AI engines say about you when your buyer goes looking.
Nothing loads from YouTube until you press play.
Plays: buyers who need you this week, with the proof
A play watches one source of buying intent and looks for the people behind it. Each one it finds arrives with one sentence saying why they are relevant and a link to the page that shows it. You approve or skip, and Scout then tells you which play started real conversations.
Relevant because: Named as a customer of Acme in their case study "How Northwind cut onboarding time".
“Northwind Freight cut onboarding time by 40% in one quarter.”
Companies a competitor names as customers. Teams hiring for the role you sell to. Fresh funding. People asking in public for a tool like yours. Visitors to your own site, contacts who changed jobs, or a list of post engagers you upload. Give Scout your website and it suggests which to run.
Every candidate carries one plain sentence - named as a customer in this case study, hiring for this role, asked for an alternative in this thread - with a link to that page and the line it quoted. When the source is your own data, such as your website visitors, it says so instead.
Each play is judged by what happened next: found, approved, contacted, replied, replied positively. Put your effort into the play that starts conversations and pause the ones that do not.
What Plays does not do: log in to your LinkedIn or X account, automate anyone's profile, or work from bought lists. It reads public pages and your own data, and nothing becomes a lead until you approve it, unless you tell a play to approve on its own.
Start with your websiteRun lead generation for every client from one place
If you source leads for other companies, each client gets its own profile, targets and lead list, and a report they can open without an account.
Each client gets its own ideal-customer profile, searches, plays, campaigns and lead lists, inside one Scout workspace your team shares.
Set how many leads a client gets each month and see at a glance who is on track and who is behind, from the day you start.
Assign or move leads between clients in bulk, and see the ones that still need a home in a single queue.
Share a read-only page with the leads delivered, which have a verified email, and progress against target. Contact details stay with you, and the link can be switched off any time.
Four ways outbound quietly stops working
Each one has a mechanism behind it, not a feature name.
Exported lists rot at roughly 2% a month. Send into them and your reputation goes down with them.
If you bought the list, so did your competitors. The prospect has read your email four times already.
So it does not happen, and the template goes out instead, and reply rates keep sliding.
Whatever the AI says next is now part of your funnel, and almost nobody can tell you what it said.
Your email lands. Then they ask an AI whether you're any good.
This is the thing now sold as Answer Engine Optimisation and Generative Engine Optimisation. Buyers no longer arrive from ten blue links; they ask an assistant and act on one answer. If that answer names three competitors and not you, the deal ended before anyone opened your email.
Prospecting tools optimise the sending and know nothing about that moment. AEO tools measure that moment and know nothing about who you contacted. Scout is the only place both sit on one schema.
- ✓ Track the questions your buyers actually ask, per engine - Gemini and Llama today, more as they are connected.
- ✓ See which rivals own the answer, and which questions are winnable.
- ✓ Raw answers stored verbatim, so a number can always be traced back to the text it came from.
An AI answer is a sample, not a measurement. Ask twice and you get different brands in a different order. Most tools run a prompt once, find you missing, and report that visibility collapsed.
Scout reports a rate only with its confidence interval and sample size, calls a change real only when the intervals separate, and excludes refusals instead of counting them as absence.
It will tell you "not enough data yet" rather than be confidently wrong. That is the point.
Four steps, and none of them are a black box
AEO and GEO are mostly sold as magic. Here is the whole mechanism, in the order it runs, so you can judge whether the number at the end deserves your trust.
Scout reads your brand, your rivals and your ICP and writes the category, comparison, alternative, problem and evaluation questions a buyer actually types. None of them name you, because a question that names you guarantees you appear.
The same questions go to each AI engine you have configured, on a schedule, many times over. One answer is a sample, not a measurement: ask twice and you get different brands in a different order.
Scout parses who was named, in what position, and which links the answer mentioned. Aliases resolve to one brand, overlapping names are not double counted, and a refusal is told apart from an answer that simply left you out.
Mention rate and share of voice per engine with a confidence interval, who owns the answers you are missing from, and which questions are winnable - ranked so a rival holding a slot beats a question nobody wins.
- ✓Tells you whether AI engines name you for the questions your buyers ask, with a number you can defend.
- ✓Names the rivals who own the answers you are absent from, including ones you never thought to track.
- ✓Records any links or publications an answer mentions. The models answer from what they learned, not a live web search, so treat these as hints about what shaped the answer rather than a citation trail.
- ✓Ranks the gaps by winnability, so effort goes where a slot demonstrably exists.
- ✓Keeps every raw answer verbatim, so any number traces back to the text it came from and history can be recomputed.
- —Write your content or publish it for you. Scout tells you what to go and earn; earning it is still work.
- —Promise a ranking. Nobody can, and anyone selling AEO or GEO guarantees is selling you variance.
- —Report a rate before it has enough usable answers to mean anything. Below the threshold it says what it still needs.
- —Count a refusal as you being absent, which is how other tools manufacture a collapse that never happened.
A tool that cannot say what it will not do is not measuring anything.
It writes the questions, not just the emails
Three places Scout does work a person would otherwise do badly, or skip. Each one ships with the guardrail that keeps it honest.
Most tools hand you an empty box and let you track your own brand name, which you always win and no buyer ever asks. Scout reads your brand, rivals and ICP and writes the category, comparison and problem questions a real buyer would type.
Each answer is parsed for who was named, in what order, and who was linked. Aliases resolve to one brand and overlapping names are not double counted, so share of voice is computed over mention slots rather than raw string hits.
Account research first, draft second. Variants are A/B tested and the winner is chosen when the intervals separate, not when one is briefly ahead.
A generated set, reviewed before anything is stored. If the model returns too little to be a set, Scout says so and falls back rather than passing a thin result off as generated.
- What is the best B2B lead generation platform for small sales teams?category
- What are the top alternatives to Apollo?alternative
- Apollo vs Clay: which is better for outbound in 2026?comparison
- How do teams usually deal with lead lists going stale?problem
- What should I look for when choosing a prospecting tool?evaluation
Each category owns half the problem
Prospecting tools know who you contacted. Visibility tools know what the AI said. Neither holds both, which is why nobody can tell you whether being cited actually converts.
| Capability | Scout | Prospecting tools | AI visibility tools | SEO / keyword tools |
|---|---|---|---|---|
| Finds and verifies buyers from live sources | ✓ | ✓ | — | — |
| Writes outreach from real account research | ✓ | ✓ | — | — |
| Measures what AI answers say about you | ✓ | — | ✓ | varies |
| Reports per engine, not one blended number | ✓ | — | varies | — |
| Confidence intervals and sample sizes on every rate | ✓ | — | varies | — |
| Keeps raw answers so past data can be recomputed | ✓ | — | varies | — |
| Writes the tracked questions from your own ICP | ✓ | — | — | — |
| Outbound outcomes and AI answers in one database | ✓ | — | — | — |
| Every found lead comes with a checkable reason (plays) | ✓ | — | — | — |
| A client area and shareable report per client | ✓ | varies | — | — |
| Syncs leads to your CRM | ✓ | ✓ | varies | — |
Columns are categories, not specific vendors, and "varies" means exactly that. Individual products differ, so check the one you are comparing against.
The questions people actually ask
Including the one where the honest answer is no.
What does Scout do?
Scout finds and verifies B2B buyers from live sources, writes outreach from what it actually found about the account, and separately measures what AI engines answer when those buyers research you. Both halves report into one database, so you can see whether being named by an AI engine turns into replies.
How is Scout different from Apollo, Clay or Lusha?
Two ways. Those tools sell you access to a contact database; Scout searches live sources, can verify each address before you send, and never sends to one it knows is invalid. And none of them measure what AI engines say about you, which is increasingly where buyers form an opinion before they ever reply.
What is a play in Scout?
A play is a saved recipe that finds the people who need your product this week from one source of buying intent, such as a competitor's published customers, an open job posting, a funding round or a public request for a tool like yours. Each person arrives with a one-sentence reason and the proof, and waits for you to approve or skip.
Where do the people in a play come from?
From public pages and your own data. Scout reads published customer pages, public job postings, funding news and public posts, and can use your website visitors, your existing contacts and lists you upload. It buys no contact lists. Each candidate says where its reason came from, and a source that could not be searched says so - it is never reported as nobody found.
Does Scout log in to my LinkedIn or X account?
No. Scout never asks for your LinkedIn or X password, does not act as you on either site and does not automate anyone's account. Plays read pages that are public, the same ones anyone can open without signing in, and when a site refuses a page Scout reports that instead of working around it.
What are AEO and GEO?
Answer Engine Optimisation and Generative Engine Optimisation are the practice of being named by AI assistants when someone asks about your category. Scout measures it: it writes the questions a real buyer would type, asks each connected engine (Gemini and Llama today, more as they are connected) repeatedly on a schedule, and reports how often you appear, per engine, with the sample size.
Can you guarantee my brand will be recommended by an AI assistant?
No, and anyone who guarantees that is selling you variance. Ask any engine the same question twice and you get different brands in a different order. What Scout can tell you is which questions you are absent from, who currently owns those answers, and which gaps are realistically winnable.
Where does the lead data come from?
Live web search plus licensed data providers where you have connected them, followed by independent email verification. Scout does not resell a static contact database, because exported lists decay at roughly two percent a month and everyone who bought the list is emailing the same people.
Will sending through Scout damage my email domain?
Scout is built to prevent that. Addresses can be verified before they are used, ones known to be invalid are never sent to, each mailbox carries a health score, and a circuit breaker stops a campaign automatically if bounce rates climb mid-run.
What happens when Scout does not have enough data to answer?
It says so. Rates are withheld below the sample size needed to mean anything, week-on-week changes are not reported when the confidence intervals overlap, and a provider that failed to respond is reported as a failure rather than as an empty result. A number you cannot defend is worse than no number.
Is there a free plan?
Yes. The free plan sources what it can from the open web at no cost and needs no credit card. Paid plans add licensed provider data and verification volume, and are set up with you so the provider budget matches what you actually send.
Can AI agents use Scout directly?
Yes. Scout has a REST API with API keys, and an MCP server is available on request, so an agent can search, enrich, verify and send using the same tools a human rep uses, with the same quotas and the same safeguards.
Can I run Scout for several clients?
Yes. Each client gets its own ideal-customer profile, searches, plays, campaigns and lead lists inside your workspace, a monthly lead target with pace, and a read-only report link the client can open without an account. Contact details stay with you, and you can switch the link off at any time.
Can Scout find and verify email addresses?
Yes. Give it a name and a company and it looks for the work email, then checks it before it can reach a sequence. Risky and catch-all addresses are labelled as such, and Scout never sends to an address it knows is invalid.
Which tools does Scout connect to?
HubSpot, Pipedrive, Zoho and Google Sheets directly, with optional automatic sync, plus signed webhooks for Zapier, Make or your own systems. Leads also import and export as CSV, and everything is available through the REST API.
How does Scout know when someone replies?
Forward replies from your inbox, or through Zapier, Make or Gmail forwarding, to your workspace's private reply address, or mark a reply by hand. A reply stops that person's sequence and is sorted as interested, not interested, out of office, referral or unsubscribe.
How is my workspace data protected?
Each workspace's data is isolated from every other. Members have roles, sign-in supports two-factor codes, API keys can be read-only and expire, and a security log records sign-ins and sensitive changes. Owners can export everything or delete the workspace themselves.
From "who do we want" to "sent" in minutes
Titles, industries, locations, company size, keywords, or plain language. Scout turns it into a live search.
Results are enriched, checked against what Scout can verify, and scored against your ICP as they reach your list.
Send from the app or hand the tools to your agent, then see what the AI engines tell buyers who go looking.
Everything a modern pipeline needs
From the first search to the reply and the CRM, in one workspace your team and your agents share.
Search live company and people data by title, industry, location, company size and keywords, instead of buying a list that is already stale.
Feed in your best accounts, then let outcomes reshape the profile as real replies come in.
Drafts built on what Scout found about the account, sequenced and A/B tested with a statistical winner, not a hunch.
Turn anonymous traffic into named companies and route hot visitors straight into the pipeline.
Funding, hiring, leadership changes and news, watched continuously so timing stops being luck.
Track how AI models describe you when a buyer researches you - Gemini and Llama today, more engines as they are connected - per model, with the raw answers kept.
A gradual volume ramp for new senders, per-mailbox health scoring and an automatic stop before a bad run damages your domain.
A full REST API, plus an agent console where you ask for prospects in plain words (an MCP server is available on request), so agents use the same tools your reps do.
Name a company and get the people who decide, by role, with their work email found and checked. Risky and catch-all addresses are marked, never passed off as safe.
Save a search or an ICP and Scout runs it on a schedule, adding only people you do not already have to the list you choose.
Know when a contact who knows you moves to a new company, so the next deal starts with someone who already trusts you.
Email steps alongside LinkedIn connect and message steps, calls and custom tasks, plus WhatsApp through your own WhatsApp Business account.
A reply stops the sequence and is sorted as interested, not interested, out of office, referral or unsubscribe. Forward replies from any inbox to your workspace's private address.
HubSpot, Pipedrive, Zoho, Google Sheets and signed webhooks for Zapier or Make, with optional automatic sync, plus CSV import and export.
Where leads stall, which source produces replies, and which step and variant earned them, with the sample size behind every rate.
Roles and invites, two-factor sign-in, read-only and expiring API keys, a security log, a full data export and self-serve workspace deletion.
Simple pricing, real leads
Every plan sources what it can for free first - you only pay for verified, provider-backed leads. Paid plans are set up with you, so your provider budget matches the volume you actually send.
Try it with zero-cost, web-sourced leads.
- ✓50 leads/mo
- ✓Free web-sourced leads only
- ✓100 email verifications/mo
- ✓100 outbound emails/mo
- ✓1 seat · 1 campaign
For a founder or small team running their first campaigns.
- ✓1,500 leads/mo
- ✓150 verified premium leads/mo
- ✓500 email verifications/mo
- ✓2,000 outbound emails/mo
- ✓2 seats · 3 campaigns
For a sales team that needs verified emails at real volume.
- ✓6,000 leads/mo
- ✓500 verified premium leads/mo
- ✓1,500 email verifications/mo
- ✓8,000 outbound emails/mo
- ✓5 seats · 15 campaigns
For revenue teams scaling outbound across the org.
- ✓20,000 leads/mo
- ✓1,500 verified premium leads/mo
- ✓5,000 email verifications/mo
- ✓40,000 outbound emails/mo
- ✓15 seats · 100 campaigns
For companies that need volume, seats, and a dedicated line.
- ✓60,000 leads/mo
- ✓5,000 verified premium leads/mo
- ✓15,000 email verifications/mo
- ✓120,000 outbound emails/mo
- ✓50 seats · 1000 campaigns
Prices in USD. Paid plans are set up with you on request, and every one is priced for real provider costs at real usage - no surprise overages.
Give your agents a pipeline of their own
Connect through the REST API (an MCP server is available on request) and let your AI agents search, enrich, and reach prospects with the same tools your reps use.