Case Study: iPEC Coaching, 2025–2026

A Campaign Architecture Built to Scale: Proven Across Two Verticals

The same demand generation framework ran at two organizations in two years: Uplift Education, a 23,000-student charter school network, and iPEC, one of the largest ICF-accredited coaching certification companies in the world, 25,000+ graduates generating $500M+ in coaching revenue. Two different verticals, the same architecture, both built from nothing.

6,913
MQLs Generated
1,557
Meetings Booked
316
Deals Won (Q3)
5.95x
ROAS Post-Cut
The short version
What I built

A full demand engine from nothing: HubSpot automation, intent-based scoring, multi-channel acquisition across three revenue streams, and a B2B intent pipeline built with n8n and ZoomInfo. No inherited infrastructure.

What it produced

6,913 MQLs, 1,557 meetings, 316 deals, 5.95x ROAS — all maintained through a 58% mid-year budget cut. The same architecture ran at Uplift Education the year before with the same results.

Why it's relevant

The same architecture ran at two organizations in two different verticals, coaching certification and K–12 education, both built from nothing. The trigger logic, scoring model, sales alignment, and spend discipline aren't tied to what's being sold. They transfer to any considered-purchase funnel.

What I Inherited

iPEC had strong product, an engaged community, and a $3.3M enrollment goal. Marketing infrastructure to hit it: none.

State on Day One

  • No automation infrastructure: every campaign was manual
  • No intent-based segmentation: everyone got the same message
  • Marketing and sales operated in separate lanes
  • Pipeline tracked by MQL volume, not opportunity creation
  • Vague positioning, no repeatable campaign framework

What I Was Asked to Build

  • End-to-end demand generation program for coaching certification
  • Scalable HubSpot automation with behavioral scoring
  • Multi-channel acquisition across paid, organic, and AI pilots
  • Weekly pipeline reporting tied to sales and RevOps
  • 4-person team aligned to pipeline contribution, not vanity metrics

This wasn't the first build. The year before iPEC, I ran the same playbook at Uplift Education, a 23,000-student, 20-campus charter school network with a $20M+ per-pupil funding target tied to enrollment. HubSpot automation built from scratch, behavior-based lead scoring, dynamic audience segmentation. Result: 4,000+ enrollment applications and 75% student re-enrollment rate. Different product, same system.

Three Revenue Streams. One Team. Concurrent.

The iPEC marketing function wasn't one campaign type. It was three distinct businesses running in parallel, each with separate buyers, offer structures, sales motions, and campaign cadences.

B2C Enrollment

CTP Certification

High-consideration, long-cycle enrollment product. Buyers research 3–6 months before committing. Webinar-anchored campaigns, behavioral lead scoring, admissions handoff, and multi-touch attribution across 3–4 cycles per year.

Program value~$12K per student
Annual enrollment goal$3.3M
Campaign cycles per year3–4
Primary channelPaid search + email sequences
eCommerce

ELI Assessments

Direct-to-coach assessment tools sold via Shopify-style checkout. Two SKUs (Self $60, 360 $119) plus two courses ($210/$299). Promotional flash sales layered onto the enrollment calendar, each with dedicated email, social, and landing page campaigns.

Jan–Sep 2025 gross sales$601,482
Orders in period2,065
Units sold9,546
Sale campaigns per year3 cycles
B2B / Corporate

Conscious Leadership

Corporate workshop program targeting HR leaders and L&D buyers. $1,500 live virtual sessions for 12–50 attendees. Supported with a dedicated sell sheet, B2B landing page on ipeccoaching.com, and a dedicated sales rep pipeline. Separate buyer persona, separate campaign motion.

Workshop price$1,500 per session
Target audienceHR, L&D, team leads
Assets builtSell sheet, 1-pager, landing page
Sales motionDedicated B2B rep + marketing support

Three Layers. One System.

Every piece of the campaign was designed to trigger from behavior, not schedules, not gut feel.

Layer 01

Audience Architecture

Three behavioral segments. Three distinct nurture tracks. Automatically routed by HubSpot based on real interaction signals.

  • Engaged (opened + clicked): resource sequences + next step CTAs
  • Hesitant (opened, no click): value prop + FAQ + social proof
  • Unaware (no open): subject line A/B + incentive + feedback loop
Layer 02

Intent-Based Infrastructure

AI-assisted lead scoring tied to behavioral thresholds, not time-in-funnel. Score hit means sequence enrollment and BDR task, simultaneously.

  • Page visit patterns, content downloads, email engagement as signals
  • Trigger-based nurture: threshold hit: sequence + BDR task
  • Multi-touch attribution across every stage to pipeline
  • Scaled to 4,918 active contacts at 32.4% open rate
Layer 03

Multi-Channel Activation

Five channels running in parallel, all tied to meetings booked and deals won, not cost per click.

  • Google Ads: 6.3% blended CTR, 920 conversions, $66.65 avg CPA (Q1 2026, account-level)
  • Waste rate held to 4.1% (industry avg: 10%) via weekly negative keyword audits
  • Meta + Pinterest optimized by funnel-stage conversion, not CTR
  • AI pilots: quiz funnels + chatbot lead magnets for top-of-funnel qualification
  • Replay email: 15.74% CTR vs. 0.5% on initial invite
B2B Intent Stack: How the Sales Rep Got a Warm List

The B2B side had no inbound. Tanya was working a cold list. I built her a better one. n8n scraped job boards daily for companies posting Leadership Development or L&D Manager roles. Those postings are a reliable signal: the company is actively investing in people development, which is the same budget that funds coaching workshops. ZoomInfo enriched each hit with firmographics, org depth, and the right contact to call. When a company cleared the size and industry threshold, n8n dropped a HubSpot task with the job posting attached as context. Tanya's queue went from a spreadsheet to a prioritized list where every row had a reason to reach out.

Signal sourceJob board postings (Indeed, LinkedIn Jobs)
EnrichmentZoomInfo firmographics + decision-maker contacts
Orchestrationn8n + crawl4ai + HubSpot
OutputPrioritized HubSpot task queue with posting evidence

The Framework

A single quarter of results isn't a system. I built this to run at Uplift Education and then brought it to iPEC with minor adjustments. A new vertical or product line could be onboarded to the same architecture in a few days.

Campaign Brief Template

Vertical, use case, offer, channel mix, and measurement plan in one doc. Co-authored with sales before launch, not distributed after.

Trigger Hierarchy

Behavioral signals defined before campaigns launch, not added post-hoc. Each threshold maps to a specific sequence entry and BDR action.

Weekly Pipeline Review

Sales in the room. Pipeline contribution as the metric: not MQLs, not CTR. Every channel decision made with field signal from reps.

Post-Mortem Cadence

Which signals predicted conversion, which didn't. Reweighted scoring model after each cycle. Attribution agreed with RevOps before launch.

Elevate Your Next Chapter

Late-cycle CTP enrollment campaign, March 2026. A webinar-anchored, multi-channel push across paid search, email, social, and landing pages over a 3-week enrollment window.

Campaign Type Late-Cycle Enrollment Webinar
Window March 7 - 27, 2026 (3 weeks)
Goal CTP Enrollments, Late Cohort
Channels Active Paid Search, Email, Social, Landing Pages, YouTube
Team Running It 4-person squad (email, design, social, copy)
Paid Search
Campaign YCTJYC / LPT/CTP
Spend (90-day) $24,074
Conversions 339
Avg CPA $71.02
ROAS 3.19x
Blended CTR 6.3%
Email
Invite Delivered 19,896
Invite Open Rate 23.52%
Invite CTR 0.5%
Replay Open Rate 43.52%
Replay CTR 15.74%
Sequences Active 3 (Attended / Not-Attended / Referral)
Organic Social
Platforms LI, IG, Reddit, FB Community
YouTube Replay Published Elevate Your Next Chapter webinar
Testimonial Asset Nathan “3x investment” clip
Community Posts Late-cycle urgency posts, daily FBCC
Daily Cadence M-F engagement maintained
Influencer Program Pilot launched 3/18
EmailWebinar invite (2 versions), replay email, "What You Missed" clip email
SequencesAttended track, Not-Attended track, Ambassador referral sequence
Landing PagesReplay page (dual CTA: Watch / Talk to Admissions), late-cycle enrollment page
Paid SearchYCTJYC / LPT/CTP campaign across 7 ad groups, negative keyword list
VideoYouTube replay upload (Elevate webinar), Nathan testimonial clip, "What You Missed" short-form clip
Social7 Days of ELI FBCC series (7 posts), daily LI/IG/Reddit engagement cadence
DesignEmail thumbnails, static ad creative, video stills, replay page layout
HubSpot OpsZapier integration, HS registration workflow, active segment refresh, contact upload

Static ad creatives produced across concurrent campaigns running alongside the Elevate webinar, each mapped to a distinct audience segment and funnel stage.

5 Mindset Shifts webinar ad creative 1
5 Mindset Shifts: Webinar Series
5 Mindset Shifts webinar ad creative 2
5 Mindset Shifts: Variant 2
5 Mindset Shifts webinar ad creative 3
5 Mindset Shifts: Variant 3
AI Chat lead magnet ad
AI Chat Lead Magnet
AYMTBAC quiz funnel social ad
AYMTBAC Quiz Funnel
7 Levels of Energy webinar ad
7 Levels of Energy Webinar
CTP Power of Choice webinar ad
CTP PoC Webinar, V2

The Infrastructure, Running

The reporting layer I built to connect every channel to closed deals. Captured March 22, 2026, three weeks into the late-cycle enrollment window.

EC GM Dashboard Overview, contacts by channel and Google Ads campaigns to deals
EC: GM Dashboard Overview Contacts by channel (original source + lead source)  ·  Google Ads campaigns to deals
HubSpot Q3 2025 Highlights — iPEC Coaching campaign results
HubSpot-generated quarterly report · every headline stat confirmed by the platform

Then the budget got cut by 58%.

Mid-year. Without warning. Pipeline commitments unchanged.

The wrong response: reduce spend evenly, maintain breadth, hope for the best. That burned two weeks before I stopped it.

Step 1: Diagnose

Ran funnel conversion analysis by channel. Which dollars were producing closed deals, not MQLs, not meetings, deals.

Step 2: Concentrate

Cut Pinterest, reduced Meta, concentrated spend on Google Ads and top HubSpot sequences. Leaned into organic + partnerships as zero-cost layer.

Step 3: Tighten the Cadence

Weekly optimization reviews became non-negotiable. Every channel decision justified by pipeline data, not media habits.

Example: Paid Media QS Audit

"Coaching certification" was our highest-spend keyword at QS 3 with $8,600 in spend. Identified that improving to QS 7 would cut CPC by ~64%, saving $5,500 annually at existing volume.

5.95x
ROAS held for the full year

The Numbers

Against a $3.3M enrollment program. 4-person team. $300K annual budget, cut to $126K mid-year.

6,913
MQLs Generated
Q3 2025, across all channels
1,557
Meetings Booked
MQL-to-meeting conversion via trigger sequences
316
Deals Won
Q3 alone, coaching certification enrollments
6,130
Form Submissions
Top-of-funnel entry across all channels
32.4%
Email Open Rate
4,918 active contacts at scale
5.95x
ROAS Post Budget Cut
Full-year, post 58% mid-year reduction

Built Once. Proven Twice. Portable.

This system wasn't built for one industry. It ran at Uplift Education, a 23,000-student school network, and at iPEC, a global coaching certification company: two different verticals, the same architecture, both built from nothing. The mechanics don't care what you sell. Here's how each component transfers to a new business.

Component What I built at iPEC The transferable principle
Intent Scoring Custom intent pipeline: job board monitoring (n8n + crawl4ai) and ZoomInfo enrichment feeding scored HubSpot tasks to the sales rep, with the triggering evidence attached. Score on behavior, not time in funnel. Whatever signals predict a purchase in your business — page paths, content downloads, product usage, renewal windows — feed the same trigger model.
Multi-Stakeholder Routing Three buyer types (individual candidate, L&D sponsor, admissions decision-maker) on separate HubSpot tracks triggered by role signal. Most considered purchases involve more than one decision-maker. Each role gets its own track, sequenced by where they sit in the decision, instead of one broadcast to everyone.
Sales Handoff Logic Score threshold hit = a sales task created in HubSpot with context. No manual queue, no lag between signal and outreach. Signal hits the threshold, the system creates a sales task with context attached, automatically. No manual queue, no lag between intent and outreach, in any CRM.
Spend Decisions 58% budget cut: pulled conversion data by channel and stage, cut channels that generated MQLs but not deals, concentrated on what closed. Cut what generates activity but not revenue, concentrate on what closes. The discipline holds whether the budget is growing or gets cut 58% mid-year.
Sales Alignment Weekly pipeline reviews with Admissions: every channel justified by meetings booked and deals won, not MQL volume. Marketing and sales in the same weekly review, measured on pipeline contribution rather than vanity metrics. Every channel decision made with field signal from reps.
Content Architecture Replay outperformed invite (15.74% vs 0.5% CTR): on-demand content sequenced by behavioral signal, not a broadcast calendar. On-demand, behaviorally-sequenced content beats broadcast for any considered purchase. The right message at the right moment, triggered by what each contact actually did.
First 90 Days, Any Build
Days 1–30: Audit

Map where the funnel leaks. Identify what marketing contributes at each stage versus what sales absorbs manually. Establish an attribution baseline before touching spend.

Days 31–60: Architecture

Define the scoring model and segment tracks with sales. Map the behavioral signals that predict a purchase in this business to CRM properties. Wire trigger logic to the score thresholds.

Days 61–90: Pilot

Run one campaign end to end against a real target segment. Measure by pipeline movement and conversion, not impressions. Prove the model before scaling it.

Let's talk about what this looks like for you.

Edward Chalupa · Demand Generation & Marketing Operations · Dallas, TX