Automated Lead Prospecting System for a $100M+ Facilities Company
Allegiance Industries · 2026
A one-off system build AfM delivered in 2026. It shows the kind of system work behind the managed service, not a separate product on sale today.
Companies Researched Per Daily Run
Annual Savings (Client Estimate)
Platforms Integrated
Data Points Per Lead
Result proof path
Outcome
A daily prospecting run with no manual research, which the client estimates saves them $9,000 to $10,000 a year.
Mechanism
Google Sheets, Make.com, Relevance AI, Apollo, and Zoho CRM connected into one daily prospecting system.
Proof
Client video testimonial plus 17 mapped data fields per lead.
The Situation
Allegiance Industries is a $100M+ facility services company headquartered in Columbia, South Carolina, with 1,200+ employees operating nationwide across janitorial, security, and electrical divisions. Their sales operation depended on manual research: individual reps searched LinkedIn, company websites, and industry directories one prospect at a time, an estimated 15 to 25 hours a week on prospecting alone. The team had tried lead lists from third-party data vendors, but the quality was poor: outdated contact information, missing direct phone numbers, and no way to check whether a lead matched their ideal customer profile. With two verticals to target, Manufacturing and Education, the research load was holding back pipeline growth. Each vertical needed different decision-maker titles, company characteristics, and qualifying criteria: a rep targeting a school district needed facilities directors, while one targeting a manufacturing plant needed operations managers. They needed a way to find, research, and enrich decision-maker contacts without adding headcount, with leads ready for outreach when the sales team started their day.
The Challenge
Manual Research Bottleneck
Sales reps spent an estimated 15 to 25 hours a week researching prospects one at a time across LinkedIn, company websites, and industry directories. That set a ceiling on how many new prospects could enter the pipeline each week.
Inconsistent Lead Quality
Previous attempts with third-party lead lists produced outdated contact information, missing phone numbers, and unverified email addresses. Reps lost further hours chasing dead leads, and the inconsistency eroded trust in any prospecting data the team had not gathered themselves.
Dual-Vertical Targeting Complexity
Manufacturing and Education need different decision-maker profiles, qualifying criteria, and outreach messaging. A single prospecting process could not serve both well, so the team ran parallel manual workflows that doubled the research burden.
No CRM Automation
Every lead had to be entered into Zoho CRM by hand, with reps copying data across fields. This introduced entry errors, inconsistent formatting, and duplicate records that cluttered the pipeline and made reporting unreliable.
What Was Built
In 2026 AfM designed and deployed a lead prospecting system across five connected platforms that runs each morning without manual research. The design is modular: each platform does one job, and the AI agent in the middle makes the research decisions a person used to make. The system reads target companies from Google Sheets, the simplest interface for the sales team to add new targets. Make.com handles orchestration, triggering daily scheduled scenarios that process companies in turn and manage error handling, retries, and logging. A custom Relevance AI agent receives each company, decides which job titles to search for based on the industry vertical, builds the Apollo queries, validates the contacts returned, and assembles the CRM record. Apollo provides the contact data: email addresses, direct phone numbers, LinkedIn profiles, and company firmographics. Zoho CRM receives the enriched leads with 17 mapped fields and duplicate detection that checks existing records before creating new ones. The run starts at 8:00 AM, so the sales team begins the day with a fresh batch of researched leads. The only manual step left is adding target companies to the sheet.
Step 1
Google Sheets
The sales team adds target companies to two spreadsheets, one for Manufacturing and one for Education. Reps only enter a company name and basic details, which keeps the interface simple for non-technical users.
Step 2
Make.com
Orchestration layer that runs daily scheduled scenarios at 8:00 AM. It reads companies from both spreadsheets, passes them through the AI agent in turn, handles error recovery and retries, and logs every action so each run can be audited.
Step 3
Relevance AI Agent
The decision-making core. It receives company data, decides which job titles to search for based on vertical and company size, builds Apollo queries with the right filters, checks returned contacts against quality thresholds, and assembles the CRM record with all 17 fields.
Step 4
Apollo
Contact data layer. The agent queries the people search API with industry-specific parameters and gets back email addresses, direct phone numbers, LinkedIn profile URLs, and company firmographic data for up to three contacts per company.
Step 5
Zoho CRM
Output layer where enriched leads land with all 17 mapped fields filled in. Duplicate detection checks existing records before creating a new lead, which prevents the duplicates that the manual process used to create.
The Results
Up to 41 / day
Company Processing Capacity
Configured to process up to 41 target companies in each daily run
Up to 123 / day
Contact Capacity
Up to 3 decision-makers for each of up to 41 companies. This is configured capacity, not an observed daily average
None
Daily Manual Research
The run completes at 8:00 AM. The team only adds target companies to a sheet; the system is estimated to replace 15 to 25 hours a week of manual research
17
Data Fields Per Lead
Full profiles including name, verified email, direct phone, job title, company, LinkedIn URL, and firmographics
$9-10K
Annual Cost Savings
The client's own estimate, given in the video testimonial below
2
Industry Verticals
Manufacturing and Education, each with its own decision-maker targeting logic
How It Works Day-to-Day
The system runs on a scheduled daily cycle. Each morning at 8:00 AM, the Make.com orchestrator reads any new companies added to the Google Sheets inputs. The AI agent processes each company in turn, working out the right decision-maker titles for the vertical, querying Apollo for contacts, and checking the data before pushing it to Zoho CRM. The sales team's only job in the loop is to add target companies to a spreadsheet when they identify new prospects; the enriched leads are waiting in their CRM the next morning. Reporting is built into the orchestration layer: every run logs how many companies were processed, how many contacts were found, and any errors. If a company returns no valid contacts, it is flagged for manual review rather than silently dropped. After launch, AfM continued to support tweaks and fix issues as they arose, including a later correction to how address fields were written to the CRM.
What This Demonstrates
Verified video testimonial
The video proof is represented by the verified client quote below so this public page does not depend on a restricted external embed.
"Jacob and his team implemented an awesome AI agent for us that saves us around 9 to $10,000 per year. Instead of being limited at around 100 new leads per day, Jacob put in a system for us within a couple of days and it really just came out flawless. I was extremely shocked at how well it worked right away. Would hire him again, and most likely will for future projects."
Eric Ormsbee
Director of Inside Sales and Marketing, Allegiance Industries
See More Results
Global Technology Platform (Anonymised)
97% Lower Cost Per Lead Over About 11 Months, $110K+ Managed
Work Jakub delivered as a freelance consultant before founding AfM (August 2025)
View Result →
ToastPal
$528K Revenue Tracked from Google Ads Over 12 Months
Work Jakub delivered as a freelance consultant before founding AfM (August 2025)
View Result →
Global Technology Platform (Anonymised)
Inbound Lead Processing and Enrichment Automation
Part of the same engagement as the paid media result
View Result →
Related Service
Past AI Marketing work is shown here with its context and qualifiers. See every result on one page.
Frequently Asked Questions
How long did it take to build and deploy the system?
In the client's words, the system was put in "within a couple of days". That covered connecting all five platforms, configuring the agent's targeting logic for Manufacturing and Education, mapping all 17 CRM fields, and testing the data before going live.
Can the system be adapted for additional verticals beyond Manufacturing and Education?
Yes. The agent's targeting logic is configurable rather than hardcoded. Adding a vertical means defining the decision-maker titles, company size thresholds, and qualifying criteria for that industry, then adding a Google Sheet for it, which the orchestration layer picks up on the next daily run.
What happens if the system encounters duplicate leads already in the CRM?
Duplicate detection checks existing Zoho CRM records before creating any new lead, matching on email address, company name, and contact name. If a match is found, the existing record is kept and the duplicate is logged for review rather than overwriting any data.
Want this kind of work on your marketing?
Book a call. Tell us where the business is now, and we will tell you what running your marketing would look like: what we would manage, what Jakub reviews, what you approve and how we would measure it.
Book a Call