Back to Case Studies
Dialpad

How Did Dialpad Build a Self-Serve Marketing Engine From Broken Infrastructure?

A complete foundation overhaul transformed broken identity systems and siloed teams into a unified, self-serve lifecycle marketing engine—reducing 1200 tracked events to 22 while dramatically improving data trust.

Quick Facts

CompanyDialpad
IndustrySaaS / Business Communications
ScaleEnterprise UCaaS Platform
Timeline12 months
Tools UsedSegment, Customer.io, Pendo, BigQuery
Key Result98% event reduction + self-serve growth marketing

Key Takeaways

1

Fix identity first—when your UUID structure is broken, every downstream tool suffers. Identity resolution must precede optimization.

2

Less data beats more data when teams actually use it. Reducing 1200 events to 22 improved trust and enabled action.

3

Self-serve marketing templates unlock velocity. When non-technical marketers can build without engineering, campaigns ship faster.

Who Is Dialpad?

Dialpad is a leading cloud-based business communications platform. Their UCaaS (Unified Communications as a Service) solution powers voice, video, and messaging for distributed teams worldwide.

Like many fast-growing SaaS companies, Dialpad's marketing infrastructure had evolved organically—scrappy solutions that worked early on had become technical debt, preventing the sophisticated lifecycle marketing their scale demanded.

What Was Dialpad's Challenge?

Broken Data Foundation + Siloed Marketing Execution

The Business Problem

Dialpad wanted to implement modern lifecycle marketing, but their data infrastructure couldn't support it. Nobody trusted the data enough to build on it—and marketing couldn't execute without constant engineering support.

What Was Broken?

UUID chaos across the entire stack—inconsistent identifiers meant systems couldn't reliably track users
Rampant legacy data nobody understood—teams didn't know which events to use
No unified view of the user journey—the company had never mapped their complete customer path
Engineering bottleneck for every campaign—marketers couldn't build or iterate without dev resources
1200 tracked events with no governance—more noise than signal
Website optimization flying blind—conversion analysis wasn't tied to user behavior patterns

The infrastructure had grown organically over years. Fixing it required a complete foundation overhaul—not patches.

How Did GrowthBench Solve It?

Complete Stack Overhaul + Self-Serve Enablement

We approached this as a layered transformation—each phase building on the previous one. Identity first, then data quality, then activation.

1

UUID Overhaul & Identity Resolution

We established a standardized SHA256 UUID as the primary key across every system. This included strategic guidance for implementation in Segment and BigQuery, a comprehensive project plan breaking down tasks by team, enabling Profiles Sync to eliminate complex multi-source SQL queries, and managing rollout across engineering and vendor teams.

2

Data Tracking Optimization

We rebuilt their tracking plan from scratch—manually auditing all legacy events from Unify and Amplitude, then optimizing from 1200 events down to 22 prioritized events. Each event maps to specific Growth use cases. We created and enforced a comprehensive Protocols Tracking Plan with governance policies.

3

Lifecycle Marketing Enablement

With clean data in place, we evaluated lifecycle marketing tools, selected Customer.io, and built a complete self-serve system. This included custom subscription state management between Marketo and Customer.io, the first-ever complete user journey map, 17 detailed communication briefs, and self-serve templates for non-technical marketers.

4

Website Conversion Analysis

Using Amplitude, we uncovered systematic conversion patterns—building workbooks tracking anomalies and traffic, developing user journey maps highlighting dropoff moments, and identifying CTA performance disparities (7.2% vs 0.8% conversion). Delivered executive dashboard with optimization recommendations.

5

Destination Re-Integration & Journey Migration

With the new UUID in place, we audited all 223 Segment Audiences and Journeys, manually recreating the most valuable ones with new targeting and orchestration. We executed custom workarounds to prevent Pendo double payment overages, re-connected destinations using new Actions methods, and conducted comprehensive QA.

What Were the Results?

Self-Serve Marketing Engine + Trusted Data Foundation

98%
Event reduction
1200 → 22
223
Audiences audited
17
Lifecycle briefs
9x
Conversion gap found
7.2% vs 0.8%

Transformation Outcomes

Unified identity across all systems—one SHA256 UUID as the source of truth
First-ever complete user journey map enabling targeted lifecycle campaigns
Self-serve marketing templates—non-technical marketers now own campaigns end-to-end
Data governance via Protocols—Growth team owns their tracking taxonomy
Profiles Sync eliminates complex multi-source SQL queries
Executive dashboard with conversion analysis and optimization recommendations
💡

Key Discovery: CTA Performance Gap

Website conversion analysis revealed a 9x performance gap between CTAs—7.2% conversion vs 0.8%. This kind of insight was impossible before the foundation work. With unified identity and clean tracking, optimization opportunities become visible.

Frequently Asked Questions

Common Questions About This Project

The complete foundation overhaul took 12 months. This included UUID standardization across all systems, tracking plan optimization, lifecycle marketing platform selection and implementation, website conversion analysis, and auditing 223 existing audiences and journeys to manually recreate the most valuable ones. The phased approach ensured each layer was solid before building the next.

Struggling With Data Infrastructure Debt?

We've helped companies like Dialpad transform broken foundations into self-serve marketing engines. Let's see what's holding your team back—and how to fix it.