Lead Scraping at ScaleBothAgentic
"Scrape 1,000 leads matching this ICP" from directories, LinkedIn, Maps, niche databases.
scrapingbrowser-actionsenrichment
Every revenue-side automation worth productizing, organized by the RACE theorem — Reach strangers, Acquire their interest, Convert them to customers, Expand what they spend. The closer a system sits to the revenue event, the more a buyer will pay for it. Each entry is tagged for B2B/B2C fit, classic vs. agentic build, and the AI capabilities it consumes.
Systems that manufacture attention from strangers. The highest-leverage stage for any company whose bottleneck is pipeline volume — and the raw-material layer every other stage depends on.
"Scrape 1,000 leads matching this ICP" from directories, LinkedIn, Maps, niche databases.
Apollo/Clay-style filtered pulls from B2B contact databases by title, industry, headcount, geography, tech stack.
Google Maps / Yelp / directory harvesting for local-service targeting — name, phone, reviews, website quality.
Cascade each lead through multiple data providers until email, phone, and firmographics are found and verified.
Monitor for buying signals: funding rounds, new hires, job changes, expansions, permits filed, tech installed or removed.
Surface accounts actively researching your category — content consumption, review-site activity, search intent.
Companies hiring for role X are buyers of Y — scrape postings into a targeted lead list.
Harvest engaged followers and commenters of competitors, influencers, or niche communities into an outreach list.
Extract who publicly uses a competitor — case studies, review sites, testimonials, "powered by" footers.
Take the best existing customers, extract their shared traits, generate lists of similar companies or people.
Identify companies (B2B) or households (B2C, where lawful) visiting the site and feed them to outbound.
Dedupe, verify emails and phones, suppress existing customers and do-not-contact records before any send.
Domain purchase, DNS/SPF/DKIM/DMARC setup, inbox creation, warm-up sequencing, rotation.
Per-lead research — site, LinkedIn, news, reviews — distilled into unique first lines, angles, and P.S. hooks.
Automated A/B/n testing of subject lines, angles, and cadences with auto-promotion of winners.
Sort every reply — interested / not-now / referral / unsubscribe / OOO — and route each to the right next action.
Agent answers first-touch questions, handles soft objections, and pushes toward a booked call — no human involved.
Inbox-placement testing, bounce and spam-rate watching, auto-pausing of burned domains.
Leads who said "check back in Q3" are auto-parked and re-engaged on the promised date with full context.
AI places first-touch calls, qualifies, handles objections, and books meetings.
Auto-dialing with local caller IDs and automatic disposition logging to the CRM.
Rank who to call right now based on engagement signals — opened email twice, visited the pricing page.
Pre-recorded — or AI-generated, per-lead personalized — voicemails delivered at scale.
Sequenced text outreach with consent management, quiet hours, and opt-out handling.
Template-based business-initiated conversations — common in international B2C and B2B.
AI holds the text conversation after the first message — qualifies, answers, books.
Automated connection requests, human-like delays, and message cadences with reply detection.
Auto-like and comment on a prospect's posts for days before the first DM so the name lands familiar.
Outbound and comment-triggered DMs ("comment PRICE and we'll DM you") at scale.
Generate and post founder/rep content on a cadence so outbound lands warmer.
Monitor Reddit, Facebook Groups, Slack and Discord communities for buying-intent posts; flag and respond.
AI produces copy, image, and video variants; launches them, kills losers, scales winners.
Continuously shift spend across channels and campaigns toward the best CAC.
Auto-sync customer and lead segments into ad platforms for exclusion, lookalikes, and targeting.
Behavior-based audiences — visited pricing, abandoned cart, watched 50% of video — fed into sequenced ad journeys.
Per-campaign or per-keyword landing pages generated and split-tested automatically.
Track competitors' ad libraries and spend patterns; alert and suggest counter-angles.
Hundreds or thousands of templated pages — per city, per use case, per integration — generated and published.
Blog, newsletter, and video-script production pipelines from keyword research through publishing.
One long-form asset — podcast, webinar — auto-atomized into clips, posts, emails, and articles.
Auto-create and maintain presence on every relevant directory, review site, and marketplace.
Trigger personalized postcards and letters — even handwritten-style — from CRM events or list uploads.
Send gifts to high-value prospects at trigger moments to open doors.
Scrape attendee, exhibitor, and speaker lists; pre-book meetings before the event.
Identify and reach potential resellers, affiliates, and referral partners — a reach multiplier.
Systems that make sure no expressed interest is ever lost, slowed, or misrouted. The stat this stage exploits: responding within 1–5 minutes multiplies conversion severalfold.
Every form — site, ads, landing pages — writes a normalized lead record instantly. No spreadsheet exports.
Chat widget that answers from a knowledge base, qualifies, and captures contact info 24/7.
Any unanswered inbound call triggers an instant SMS. Huge for local service businesses.
Virtual receptionist answers every inbound call, takes the request, books or routes it. No voicemail black hole.
Comments and DMs on social auto-converted into CRM leads with the conversation attached.
Download, webinar, and quiz funnels that deliver the asset, tag the interest, and start the right follow-up track.
Per-channel phone numbers with recorded calls tied back to the campaign that produced them.
New lead triggers an email + SMS + call attempt within seconds of form submission.
Voice agent calls the lead within a minute, qualifies, and books the appointment while intent is hot.
Nights, weekends, and call spikes handled by AI so no lead waits until morning.
Instantly notify the assigned rep with lead context; auto-reassign if not claimed within N minutes.
The moment a lead arrives, append firmographics, social profiles, and history before anyone touches it.
Grade fit (ICP match) and intent (behavior) so reps work the list in value order.
Conversational budget-and-timeline qualification before a human ever engages.
Auto-generated one-page dossier on each qualified lead, delivered before the call. = Lead Researching System.
Bots, competitors, existing customers, and duplicates screened out before they pollute pipeline metrics.
Round-robin, territory, language, deal-size, or specialty-based assignment with load balancing.
Qualified leads book directly onto the right calendar; ownership and reminders handled automatically.
Confirmations, reminders, day-of texts, easy-reschedule flows, and auto-rebooking after a no-show.
When a lead moves (SDR→AE, marketing→sales), the full history and research brief move with it.
Between booking and the call: case studies, explainer video, what-to-expect — raising show rates and close rates.
Content sequences matched to the lead-magnet topic or the pages visited.
Leads that went quiet before ever talking to sales get automatically re-warmed with new angles.
"This lead just revisited the pricing page" — routed to the owner in real time.
Systems that compress the time and effort between "interested" and "paid," and stop deals dying of neglect.
Call transcript + CRM data → branded, scoped proposal in minutes instead of days. = Auto Proposal System.
Structured inputs — or photos, plans, measurements — priced instantly; includes CPQ for complex catalogs. = Auto Quote System.
Turn a happy-customer interview or project record into a polished case study — on autopilot after every win. = Case Study Creator.
Personalized value models generated per prospect to arm the internal champion.
Per-prospect walkthrough videos or demo environments seeded with their data and branding.
Shared close-plan documents generated and kept current from deal-stage data.
Rep — or agent — instantly pulls the right case study, one-pager, or battle card for the objection at hand.
Every call recorded, transcribed, summarized; pains, budget, objections, and next steps extracted. The hub for the systems below.
Extracted fields — budget, timeline, decision makers, stage — written to the CRM with no rep typing.
Commitments made on the call become drafted follow-up emails and scheduled tasks automatically.
Aggregate objections across all calls; surface the top deal-killers and the responses that win.
Every call auto-graded against the sales methodology; managers coach from data, not ride-alongs.
Every open deal has a next touch scheduled — silence from the prospect never means silence from the company.
The moment a prospect opens the proposal, the rep is alerted and a timed follow-up fires.
Different cadence and content per stage — post-demo vs. post-proposal vs. verbal-yes.
Multi-channel pattern-interrupts — SMS, video, direct mail — when email follow-up goes quiet.
A final structured "closing the file" sequence that reliably resurrects a percentage of dead deals.
The B2C close: checkout or application abandoned → timed email/SMS recovery journey with incentives.
Emails, calls, and meetings logged automatically; the CRM reflects reality without rep discipline.
Deals with no activity or a passed close date flagged and queued for action or closure.
Agent reads each open deal — calls, emails, stage history — and flags risks a pipeline report can't see.
Automated pipeline-weighted forecasts by rep, team, and period.
Structured loss reasons enforced at close-lost, aggregated into win/loss insight.
Closed-won terms → contract → e-sign → countersigned copy filed, all automatic.
Invoice or payment link fires with the signed contract; deposit chased until paid. Overlaps the Auto Invoicing System at its revenue-critical edge.
Qualified financing options presented automatically at the moment of price resistance.
The win instantly kicks off onboarding: welcome email, kickoff booking, internal notifications, project creation.
Systems that grow revenue from people who already bought — the cheapest revenue in the business, because the trust is already paid for.
Scope: E is pure expansion — upsell, cross-sell, referral, reactivation. Retention and churn-prevention systems (health scores, save flows, support automation) are deliberately excluded and form a separate future bucket.
Customer hits a plan limit or heavy-usage pattern → automated upgrade offer at the moment of need.
Success moments — result achieved, anniversary, Nth purchase — trigger a congratulate-then-upgrade sequence.
Agent maps what each account owns vs. the full catalog and ranks the gaps by likelihood to buy.
Segmented campaigns moving customers up the plan, package, or service ladder with personalized justification.
One-off buyers automatically offered recurring plans — maintenance contracts, memberships, subscriptions.
Long-tenured accounts flagged for repricing or right-sizing conversations with prepared talking points.
"Customers who bought X buy Y" applied to email, checkout, portal, and rep prompts.
Timed sequences after each purchase offering the natural next product or service.
Calendar- or age-based offers — HVAC tune-up season, tire change, annual review, warranty expiring.
Auto-assembled bundles from co-purchase patterns, offered at checkout or renewal.
Before any account touchpoint, the rep sees the top cross-sell recommendation and why.
Watch existing accounts for funding, hiring surges, new locations, new product lines → expansion conversation triggered.
When a champion moves companies, two plays fire: expand into the new company, re-secure the old account.
Agent maps sister departments, subsidiaries, and locations of existing accounts as warm expansion targets.
Business-review decks auto-built from usage, results, and whitespace data — with the expansion ask included.
Boundary note: winning back lapsed customers counts as expansion here — it monetizes the existing database, the same asset E works. = CRM / Past Customer Mining system.
Dormant past customers re-engaged with an offer via email/SMS — classically the fastest-ROI automation in local business.
Two-way AI (SMS/email) that doesn't just blast the offer — it converses dormant customers back to a booking.
Predict when each customer is due to buy again from purchase intervals; trigger outreach just before.
Timed to peak-happiness moments — great review, result delivered, NPS 9–10 — with a frictionless share mechanism.
Unique links and codes, attribution, and automatic reward fulfillment for referrers.
Post-purchase review requests routed by sentiment — happy → public review sites; unhappy → private recovery.
New 5-star reviews auto-formatted into website widgets, social posts, and ad creative.
Spot the happiest customers automatically and queue them for the Case Study Creator.
Partner onboarding, asset delivery, lead attribution, and commission payout on autopilot.
The 19 reusable skills every Agentic entry above is composed from. Build a capability once and it amortizes across every system that consumes it — a capability that feeds many RACE stages is itself high-leverage.
Extract structured records from directories, social platforms, maps, job boards — "scrape 1,000 leads."
Append emails, phones, firmographics, socials from cascading providers; verify and dedupe.
Open-ended multi-source investigation of a person or company, distilled into usable findings.
An agent operating real web UIs: form-filling, portal navigation, social actions, listing management.
Standing observers of websites, news, review sites, usage data, CRM state; fire events on change.
Recognize buying signals in monitored data — funding, hiring, page visits, community posts.
Natural phone conversations, inbound and outbound: answering, qualifying, booking, objection handling.
Two-way conversational agents on SMS, web chat, WhatsApp, and social DMs.
Read, classify, and answer email threads; drive toward a booked call or decision.
Per-recipient messages, angles, and content that reads hand-written.
Proposals, quotes, contracts, case studies, briefs, QBR decks — from structured + unstructured inputs.
Ad creatives, personalized videos, social assets, voicemail audio.
Transcribe, summarize, and extract structured facts, commitments, and objections from conversations.
Grade leads, deals, accounts, content variants; rank anything by fit or likelihood.
Detect happiness, frustration, and buying temperature in replies, reviews, and conversations.
Query and update the system of record: create leads, log activity, move stages, set fields.
Check availability, book, reschedule, remind.
Send contracts, collect signatures, fire payment links, confirm receipt.
The spine that chains capabilities into a system: triggers, branches, waits, retries, handoffs.
Classic vs. agentic is exactly whether step 03 requires judgment. Classic automations hard-code the decision; agentic ones reason. The productization opportunity is largest where step 03 used to require a skilled employee's judgment — that is the labor being replaced.
Four rules for turning this catalog into productized offers — and where the existing sellable systems already sit inside it.
R, A, and C sit directly on the revenue event — a system there produces attributable dollars, which makes it easy to price and sell. E systems are cheapest-revenue plays with the warmest proof: "we made you $X from customers you already had."
Diagnose which RACE stage is the client's constraint, then productize from that stage's list. A company drowning in leads doesn't need R — it needs A2 (speed to lead) and C3 (follow-up machinery).
Favor systems whose capability tags you've already built. Each capability — voice, call-intel, gen-docs — amortizes across every system that consumes it. Orchestration, monitoring, and crm-actions feed all four stages.
The entries worth the most are the agentic ones replacing a skilled employee's judgment — quoting, proposal scoping, lead research, deal risk. Classic plumbing is the entry point; reasoning is the moat.
Applicant Tracking (Sys 07) sits outside RACE — it's a hiring-side system. See also Anatomy of a Company for the department-level revenue-proximity map this catalog zooms into.