Your pipeline doesn’t have a lead problem—it has a filter problem. If every inquiry gets a human reply, your closers get buried under “just browsing” DMs and tire-kicker form fills. We fix that by installing AI filters that qualify, score, and route leads before they touch your team.
What’s an AI Filter?
An AI filter is a short, conversational checkpoint that runs in DMs, website chat, SMS, or forms. It asks a few targeted questions, scores the answers, and triggers the right next step—book a call, nurture, or politely disqualify. Think SDR logic, automated.
Built and managed inside Ascent (AI chat + CRM + workflows), our filters are fast to launch and easy to tune.
Why Filters > More Volume
- Protect your calendar: Only right-fit leads see booking links.
- Save human time: Reps spend time closing, not sifting.
- Improve CX: Fast answers, clear next steps, no ghosting.
- Cleaner data: Every chat updates tags, scores, and pipelines automatically.
The 4-Question Filter (our proven baseline)
- Goal/Use Case – “What’s the #1 thing you’re trying to solve?”
- Timeline/Urgency – “Are you looking to start this month or exploring options?”
- Fit Signal – one niche-specific qualifier (zip/venue, budget band, property type, etc.)
- Access/Authority – “Are you the decision maker for this?”
These four give us enough to score intent and route.
Scoring & Routing Logic (simple, ruthless)
- Hot (Score 8–10): Auto-show calendar + instant SMS/Slack alert to sales.
- Warm (5–7): Deliver proof (case study/offer explainer), propose times, follow up for 3–7 days.
- Cold (≤4): Enter a light nurture + FAQ drip, invite them to a resource, remove from sales queue.
All scoring, tags, and next steps live in Ascent Workflows so you’re not duct-taping tools.
Where We Install Filters
- Instagram/Facebook DMs: Story replies, ad DMs, bio link traffic
- Website chat & popups: Exit-intent, pricing page, contact page
- SMS replies: Missed-call text-back, inbound keyword campaigns
- Lead forms: Replace long forms with 30–60s chat-based intake
Here’s How to Implement (Step-by-Step in Ascent)
1) Choose the entry points
Pick where the filter should fire first: Instagram/Facebook DMs (story replies/ad DMs), website chat (pricing/contact pages), SMS (missed-call text-back), or a short form replacement.
2) Define the 4-question filter
Lock in: Goal, Timeline, Fit signal (niche-specific), Authority.
Example: “What are you trying to solve?” → “When do you want to start?” → niche qualifier (zip/venue/budget/industry) → “Are you the decision maker?”
3) Build the conversation in Ascent
Create a chat workflow with button replies, input validation, and friendly guardrails. Add a “Talk to a human” path and (for legal/medical) a brief disclaimer.
4) Score and tag automatically
Hot/Warm/Cold thresholds in Ascent Workflows. Apply tags (e.g., hot-intent, needs-proof, not-fit) and write the score to the contact record for reporting.
5) Route and book
- Hot: show calendar, create Opportunity (stage: Qualified), alert sales via SMS/Slack.
- Warm: send a proof pack (case study, FAQs), propose times, start 3–7 day follow-ups.
- Cold: provide a helpful resource, add to light nurture, remove from live sales queue.
6) Add follow-up & safety nets
No response? Re-prompt once, then pause. Booked? Send confirmations + reminder cadence. Missed? Auto-reschedule offers. Confusion or sensitive topic? Trigger human handoff.
7) QA before you go live
Run 10–15 test conversations across edge cases (blank answers, long texts, wrong formats). Check tags, scores, calendar links, alerts, and pipeline stage creation.
8) Launch on one channel, then expand
Start where volume is highest (often DMs or website chat). Watch completion rate, hot rate, and book rate for a week; tighten copy/thresholds; then roll to remaining channels.
9) Review weekly, optimize relentlessly
Track: filter completion %, hot rate, booked-from-hot %, show rate, time-to-first-response, and closed-won. Retrain prompts using transcripts and objection logs; adjust scoring weights and follow-up timing inside Ascent.
Assets you’ll need
Offer/pricing, 2–3 case studies or reviews, FAQs/objections, calendar availability, and any compliance language (e.g., “not legal advice”) plus secure upload links if needed.
Niche Examples (prebuilt filters we deploy)
Dentists/Med-Spa
- Qualifiers: treatment type, urgency, insurance/financing
- Route: emergency path vs. cosmetic consult
- Asset: before/after proof + new-patient offer
Home Services (HVAC/Roofing/Plumbing)
- Qualifiers: service type, zip, urgency, property type
- Route: after-hours emergency → on-call; standard → scheduler
- Asset: price bands + availability window
Real Estate
- Qualifiers: buyer/seller, budget, pre-approval, neighborhood
- Route: hot buyers → showing slots; sellers → valuation call
- Asset: property list or CMA teaser
Coaching/High-Ticket
- Qualifiers: goal, current revenue/stage, program fit, budget band
- Route: qualified → strategy call; nurture → case-study drip
- Asset: 2–3 social proofs + mini training
Law Firms (PI/Family/Criminal/Immigration/Estate)
- Qualifiers: matter type, county/venue, urgency/court date, quick conflict check
- Route: urgent → immediate callback; standard → 15–30m consult
- Asset: attorney bios/verdicts + “what to expect” one-pager
The Workflow Blueprint (inside Ascent)
- Trigger: inbound DM/chat/SMS/form submit
- AI Filter: 4-question micro-intake + entity extraction (goal, timeline, qualifier, authority)
- Score & Tag: hot/warm/cold + niche tags
- Branching Actions:
- Hot → show calendar, create opportunity (stage: Qualified), alert rep
- Warm → proof pack + soft booking CTA, 3–7 day follow-up
- Cold → light nurture, remove from sales queue
- Safeguards: if confusion/sensitive topic → human handoff; if no response → re-prompt then pause
- Analytics: response time, filter completion rate, hot rate, book rate, show rate, closed-won
Copy You Can Steal (works across channels)
Opener: “Hey! 👋 I can point you to the best next step—mind if I ask 3 quick questions?”
Booking Nudge (Hot): “Sounds like a fit. I can grab you the next 15-min slot on [Day/Time]. Want it?”
Polite Disqualify (Cold): “Based on what you shared, we may not be the best fit right now. Here’s a resource to help—and if things change, reply ‘START’ and I’ll re-open your options.”
Guardrails That Keep Filters Trustworthy
- Always offer a human option (“Talk to a specialist”)
- Be explicit about next steps (book, resource, or follow-up)
- Respect data & compliance (disclaimers for legal/medical; secure uploads)
- No endless interrogations (30–60 seconds max to qualify)
KPIs We Track Weekly
Filter completion (65–85%) · Hot rate (20–40%) · Book-from-hot (60–80%) · Show rate (70–85%) · Time-to-first-response (<60s) · Time-to-first-revenue (downward trend)
Rollout Plan (14 days)
Week 1 – Build & Connect: pick qualifiers, write prompts, wire channels, set scoring/tags
Week 2 – Launch & Tune: soft launch on one channel, tighten copy/branches, enable follow-ups, push to all channels
Common Mistakes (don’t do these)
Turning the filter into a long form · Letting everyone book your calendar · No path for warm leads · No human escape hatch · Not retraining weekly on transcripts/objections
What You Get When We Install This
- A single Ascent workspace with AI chat, filters, scoring, calendar, CRM, and workflows synced
- Prebuilt niche filters + copy
- Weekly optimization (prompts, scoring weights, follow-ups)
- Reporting that ties filters to revenue, not vanity metrics
Leave a Reply