The situation
Marco runs a marketing agency with 4 people. They handle campaigns for local businesses — dentists, gyms, restaurants, that kind of thing.
His pipeline was a Google Sheet with color-coded rows. Leads came in through the website contact form, Instagram DMs, phone calls, and the occasional referral email. Whoever saw it first would add it to the sheet. Sometimes.
The problem wasn't that Marco didn't have leads. The problem was that nobody knew what to do with them once they showed up.
A lead would fill out the contact form on Monday. Marco would see it Tuesday morning. He'd forward it to one of his team. That person would check it Wednesday, google the business, spend 15 minutes figuring out who this person even is, then send a generic "thanks for reaching out" email. By Thursday the lead had already talked to another agency.
He was also double-booking intro calls because the team was scheduling from three different calendars. Two people would claim the same lead without knowing. One lead sat untouched for 11 days because everyone assumed someone else had called.
Marco knew it was bad. He just didn't have time to fix it because he was too busy managing the mess.
What it was costing
Marco estimated he was losing 3-4 leads per month to slow follow-up alone. For an agency charging $1,500-3,000/month per client, that's $4,500-12,000 in potential monthly recurring revenue walking out the door.
His team was spending roughly 45 minutes per lead just on admin — finding out who the person is, what company they're from, checking if they've been contacted before, figuring out who should handle it.
With about 30 new leads per month, that's over 22 hours of manual admin. Almost a full work week, every month, spent on stuff that adds zero value.
What we changed
We rebuilt the front door of his business.
Website + AI chatbot. We built Marco a proper site. The chatbot sits on it and handles the first conversation — answers questions about services, pricing ranges, timelines. If someone's a serious prospect, it captures their name, business, what they need, and how urgent it is. This runs 24/7, so leads that come in at 10pm on a Sunday don't wait until Monday morning for a response.
Lead enrichment. When the chatbot captures a lead's info, we run it through an enrichment pipeline before anyone on the team even sees it. Company name goes in, and we pull back: LinkedIn profile, business website, estimated company size, industry, whether they're already running ads, and their social media presence. All of this shows up automatically in the CRM entry.
The enrichment uses a combination of public data sources and API lookups. Nothing creepy — just the stuff you'd spend 15 minutes googling anyway, done in about 3 seconds.
CRM + auto-assignment. Leads land in the CRM fully enriched, tagged by service type and region, and auto-assigned to the right team member based on simple rules Marco defined. Gym leads go to Sara because she handles all the fitness clients. Restaurant leads go to Tom. If a lead doesn't match any rule, it goes to Marco directly.
The result for the team member. Sara opens her CRM in the morning and sees: "New lead — Flex Fitness, 2 locations, running Facebook ads already, owner is Daniel, wants help with Google Ads, came in last night at 9pm, chatbot already answered his pricing questions."
Sara doesn't need to google anything. She doesn't need to check a spreadsheet or ask Marco who should handle it. She picks up the phone and says "Hey Daniel, I saw you're already running Facebook campaigns for your two locations — are you looking to add Google Ads to the same strategy or test something different?"
That's a different conversation than "Hi, thanks for reaching out, how can we help you?"
The numbers
- Lead response time: from 1-2 days down to under 5 minutes (chatbot handles first touch instantly)
- Admin time per lead: from ~45 minutes to ~5 minutes (enrichment + auto-assignment eliminates the research and routing)
- Monthly admin hours recovered: roughly 20 hours back across the team
- No more double-bookings — single CRM with clear ownership per lead
- Build cost: approximately $2,800 for the full setup (website + chatbot + CRM + enrichment pipeline)
- Ongoing cost: about $150/month for the enrichment API and chatbot hosting
Marco stopped losing leads to slow follow-up. His team stopped wasting half their week on admin that a system should be handling. And every intro call now starts with context instead of cold small talk.
Need a practical read on your workflow?
Translate this into execution
Get a tailored implementation roadmap with clear milestones and ownership.
FAQ
How long did this take to build?
About three weeks from first call to everything running. The website and chatbot took the first week. CRM setup and enrichment pipeline took the second. Third week was testing, training the team, and tuning the chatbot's responses.
Does the chatbot replace a real person?
No. It handles the first touch — answers common questions and captures info so a real person can follow up with context. It's not pretending to be human. It's making sure leads don't sit in a queue overnight.
What happens if the enrichment can't find anything?
The lead still lands in the CRM with whatever the chatbot captured. The team member just has less context and might need to do a quick lookup. Works fine — it's still faster than the old spreadsheet flow.
What if my business gets fewer than 30 leads per month?
The math still works. Even at 10 leads per month, you're saving 6-7 hours of admin time and responding instantly instead of next-day. The build cost is the same — the ROI timeline just shifts from weeks to a couple months.