Every question, answered
How agents work, what's in each plan, how MCP wires up, how DFY engagements run. Skim by topic, or search anywhere on the page.
Pick a topic to read
Switch between topics without leaving the page. Your selection updates the URL so you can share a direct link.
About CatchIntent
What we are, who we serve, how we are different.
CatchIntent is the intent layer for B2B outbound. Our agents find buyers showing real timing signals — job changes, funding rounds, hiring spikes, competitor engagement, and category discussions — score them by warmth, draft personalised openers, and deliver them as warm leads every day. The unified flow: Product → Audience → Accounts → Leads → Outreach.
Primarily outbound, lead-gen, GTM, demand-gen, and fractional SDR agencies running outbound across multiple client brands. Secondarily Seed to Series A B2B SaaS teams with an SDR, sales co-founder, or founder dedicating real time to outbound. We are deliberately not built for solo founders without outbound labour, marketing-only brand-monitoring teams, or recruiting use cases.
Contact databases give you static lists. CatchIntent gives you <strong>timing</strong>. The leads in your queue are people who just did something that predicts a deal — left a job, raised a round, hired five SDRs — not just people who match an ICP filter. We layer ICP filtering, AI scoring, and drafted openers on top of that timing data.
Reddit, HN, and Bluesky monitoring belong to the older social-listening category — useful for brand monitoring, weak for B2B outbound because the authors are pseudonymous. CatchIntent is identity-first: every lead resolves to a real person on LinkedIn with role, company, and timing context your SDR can act on the same day.
No. Reddit and Hacker News authors are structurally pseudonymous and cannot be resolved to LinkedIn profiles reliably. Anyone claiming otherwise is misleading you. Our LinkedIn intelligence is sourced from public LinkedIn signals directly — not by bridging pseudonymous social.
Agents & leads
How agents work and what shows up in your queue.
Each Product has one Audience — your lead-discovery config covering ICP, competitors, keywords, value props, target roles, and which timing signals are on. Growth includes 1 Product (one Audience); Scale includes 3 Products (one Audience each).
A-tier (active job-change and funding signals), B-tier (hiring signals), C-tier (competitor engagers, LinkedIn keyword discussions), and a baseline ICP search to fill volume. Each signal type is configurable per agent.
Agents run every few hours and bulk-save leads as they are discovered. AI scoring runs within 1-5 seconds of save in normal queue conditions, so by your first standup the day's queue is enriched, scored, and ready.
Each lead has a person (name, role, LinkedIn URL), a company (name, domain, size, industry), the signal that surfaced them (e.g. "joined as VP Sales 3 days ago"), a warmth score (Hot / Warm / Cool), a one-line reason, and an AI-drafted opener you can copy, edit, or send through outreach.
Depends on plan and ICP breadth. Growth caps at 1,000 leads/month (~33/day), Scale at 4,000/month (~130/day). Enterprise is custom. DFY operators tune signal mix weekly to keep volume predictable.
Yes. Workspaces are multi-Product by default. Each Product has its own Audience, ranked book of accounts, and lead queue, so an agency or multi-product team keeps them isolated. Growth = 1 Product, Scale = 3 Products, Enterprise = custom.
No. A lead is a person showing a real reason to buy right now, not a closed deal. The number that matters is the comparison. A cold list converts at roughly 1-2% because you are interrupting people who never asked. Reaching someone the week they changed jobs, raised funding, or started asking about your category lands very differently, so reply and conversion rates are multiples higher. How much higher depends on your offer, your follow-through, and your fit. We get you in front of in-market people. Closing them is still your product and your pitch. Plan for 1,000 well-timed conversations, not 1,000 signed contracts.
It is deliberately small. This is not a static database of 100,000 contacts you blast once and burn. Every lead is a specific person showing intent that week, refreshed daily. Think of it like food: offer a meal to someone who just said they are hungry and they thank you; push the same meal on someone who is not hungry and they resent you. A giant cold list is the second case at scale, which is why it converts at 1-2% and wrecks your sender reputation. A steady stream of people actively signalling need is the first case, and a few hundred of those a month beat tens of thousands of cold contacts. You can scale volume on a higher plan once the queue proves itself. You cannot manufacture intent that is not there.
Outreach & openers
How AI-drafted messages and CRM handoff work.
No, and that's deliberate. We draft the opener, ground it in the signal, and hand it off. You send from your own LinkedIn or email account through your existing tools — Lemlist, Instantly, Smartlead, or directly. Sending from a deliverability-managed account beats sending from a vendor tool.
Each opener uses your brand voice (set during onboarding), the specific signal that surfaced the lead, and the lead's role + company context. Drafts are short (under 280 chars), specific, and editable. They cite the signal so you sound like a human who noticed, not a robot.
HubSpot, Close, Pipedrive, Apollo (as CRM), Zoho, Lemlist, and Instantly are wired up. If a workspace has multiple destinations connected, push fans out across all of them. Custom CRMs work via webhook or MCP.
A LinkedIn extension is on the near-term roadmap for one-click outreach from the LinkedIn surface itself. For now, lead handoff is via copy-to-clipboard or CRM push.
MCP & AI integrations
Operate CatchIntent from Claude, Cursor, or any MCP client.
Model Context Protocol is an open standard that lets AI agents call typed tools on external services. CatchIntent ships an MCP server so your Claude, Cursor, or Codex agent can operate on your leads, agents, outreach, and workspace through natural language — no custom integration, no chatbot wrapper.
Any spec-compliant MCP client. Tested with Claude Code, Claude Desktop, Codex CLI, Cursor, Gemini CLI, Windsurf, ChatGPT (via connectors), GitHub Copilot, Raycast, and Antigravity.
Yes. MCP access is included on Growth, Scale, Enterprise, and the 7-day free trial. There is no separate MCP add-on.
One-time OAuth 2.1 flow. Your MCP client opens the dashboard in a browser, you pick a workspace, review the requested scopes, and approve. Tokens refresh automatically. You can revoke any client any time from Settings.
Yes. The OAuth consent step shows nine scopes (read/write pairs for leads, agents, workspace, outreach, plus write:crm). Grant only what you need. Every tool checks its scope before running.
Pricing & trial
Plans, trial, what is included.
Growth ($99/mo): 1,000 leads/month, 1 Product, twice-daily refresh. Scale ($249/mo): 4,000 leads/month, 3 Products, 4× daily refresh. Enterprise (custom): higher volume, SSO, white-label, dedicated support. DFY starts at $1,999/mo with a 3-month minimum.
Yes. Every plan includes a 7-day trial with a card on file. You're not charged during the trial. Cancel before day 7 and you won't be charged.
No. Every plan starts with a 7-day trial. We chose a card-on-file trial over a free tier so we attract teams who are serious about outbound, not casual browsers.
Yes. Upgrades take effect immediately and prorate. Downgrades take effect at the next billing cycle so you keep the volume you already paid for.
Self-serve is monthly only. Annual billing is available on Enterprise and DFY engagements where it makes sense for the commitment shape.
Kelviq handles billing — credit/debit card on file with most major networks supported. Enterprise/DFY can be invoiced.
Done-for-you (DFY)
How operator-led engagements run.
DFY is operator-led outbound. A CatchIntent operator runs your brand audit, configures your agents, tunes signal mix weekly, and drafts every opener. You review the queue and send. Starts at $1,999/mo with a 3-month minimum.
Outbound needs at least one full cycle to show signal. Week one is brand audit, week two is agent setup, by week three the queue is delivering daily. The 3-month minimum gives the engagement a fair window to prove out before you decide to renew, expand, or move to self-serve.
A CatchIntent operator with B2B outbound experience. You meet them on the scoping call, they own the engagement end-to-end (audit, setup, weekly tuning, drafted openers, weekly recap), and you get a Monday recap and optional Friday alignment call.
Daily lead delivery + drafted openers. Monday: client-ready recap PDF in your inbox. Wednesday: operator-led mid-week tuning. Every other Friday: optional 30-min alignment call with your operator.
Yes. DFY engagements convert cleanly — we hand over the agent configs, signal mix, and opener templates. Most clients stay on DFY because the operational lift is exactly what they wanted to outsource, but the option is always open.
Yes. The Monday recap PDF can carry your agency's logo and colors. Your clients see your brand, not ours. Full white-label dashboard access is on Enterprise.
Agencies & multi-brand
Workspace isolation and partner programs.
Each Product inside a workspace has its own Audience, leads, integrations, and reporting. No cross-contamination between client lanes, no manual workspace switching for the operator.
Yes. Members are scoped to the workspace. Per-brand access controls are on Enterprise. Most agencies run with two or three operators per workspace, each owning a subset of brands.
Yes — perpetual 20% commission on every client you bring + a flat agency fee. Email us through the contact page to discuss. We're selective about who we partner with so the program stays high-signal.
Yes, especially on the partner program. We provide the data + drafts; you provide the operator labour and client relationship. Both pricing structures (resell, white-label DFY, side-by-side referral) work — depends on what fits your engagement shape.
Data & security
Sources, OAuth, retention, and compliance.
Public LinkedIn signals (job change, hiring, funding, post engagement, keyword discussions) sourced through licensed providers, supplemented by company-event data and (where the lead allows) public profile enrichment. We do not scrape behind logins.
OAuth tokens are encrypted at rest and scoped per workspace. API keys for tools that do not support OAuth are encrypted the same way. You can revoke any integration from Settings.
AI scoring uses our own provider (currently AWS Bedrock with Anthropic Haiku) and only the minimum context needed to score and draft. Your workspace data is never used to train external models.
SSO (SAML/OIDC) and SCIM provisioning are Enterprise features. Workspace-level audit logs are available on Scale and above.
Production runs on Hetzner (EU) and AWS for AI compute. Cloudflare fronts the marketing surface. We can discuss regional pinning on Enterprise.
GDPR-aligned by design (data minimisation, right to erase, EU hosting). SOC 2 is on the roadmap — we'll publish the report when it's ready.
Setup & onboarding
Getting started, time to first lead.
Self-serve onboarding takes ~5 minutes: brand context → ICP filters → optional offer voice → first agent enabled. Leads start flowing within hours of agent activation.
A workspace name, your brand (name, domain, one-line description), and the ICP filters you care about (target roles, industries, locations, company sizes). Everything else can be filled in later.
You need at least someone who acts on leads — a founder, sales hire, fractional SDR, or agency operator. CatchIntent is the discovery layer, not the labour. Without someone working the queue, the leads sit.
Today CatchIntent generates new leads from signals rather than scoring an imported list. Importing is on the roadmap — for now, push your CatchIntent leads into your CRM and use your existing list workflows there.
Support & contact
Where to ask, how fast we reply.
Email <a href="mailto:[email protected]" class="text-[var(--mk-accent)] underline">[email protected]</a> or use the contact page. Founder responds personally to most inbound while the team is small.
Yes — Scale and Enterprise plans get priority response (next business day or faster). DFY engagements get a direct Slack channel with your operator.
Email <a href="mailto:[email protected]" class="text-[var(--mk-accent)] underline">[email protected]</a> with steps to reproduce and a screenshot if possible. Bug fixes ship usually within a day or two of a credible report.
Yes. Feature requests with a real use case ("I would use this to do X because Y") get a real reply and often end up on the roadmap. Vague wish-lists less so.
Question we did not cover? Just ask.
Founder reads every inbound. Most questions get answered the same day.
No bot replies. No support tier maze.