Who is Cloudset?

One of Zendesk’s first implementation and apps partners

• Cloudset specializes in making complex processes and technical functions easy to understand and simple to achieve. They started back in 2009, providing Zendesk best practice implementation services.
• Cloudset optimizes the Customer Support teams’ usage of their Zendesk. They use Zendesk to develop add-on apps that enable sophisticated support solutions.

The Context

Cloudset’s Founder Graham had one big goal: Having his dream qualified, accurate Database of companies using a specific customer support platform. All above 200+ employees, across US, UK, and the main countries in Europe.

In the past, Graham tried multiple technographic sources, from Builtwith to Apollo, always finding outdated, incomplete data. Some of the companies that should allegedly use the support platform were in reality not using it, as well as many companies using the tool were missed.

In Graham’s words: “The main motivation was ABM influenced. It was not only inefficient to include out of target customers, it’s a contribution towards not spamming people. We have done our research and we have now a legitimate interest”.

Business Development Blockers

Lack of Sales Tech Know HowThe team missed the latest sales AI techniques.The team had no way to ensure the accuracy of the Technographic data, as well as finding tech-related companies beyond the ones found in Apollo or Builtwith.

Typical Databases were not enough

Cloudset tried all possible databases you could think of for technographic data:

• Apollo
• Builtwith
• Sales Navigator

All of them provided inaccurate or incomplete information, that left the Founder Graham unhappy and frustrated with the results.

How Cloudset met TAM Acceleration

Cloudset’s Founder Graham found TAM Founder Matteo’s profile featured as an Expert in La Growth Machine’s Advisor page, in July 2024.

After booking a call with Matteo, Graham decided TAM Acceleration was the right partner to come up with the Database he had always dreamt of.

Initial Goals

• Use the latest AI Techniques to ensure accuracy of the Customer Support Platform
• Merge multiple Databases together, to find as many Enterprise as possible that were using the the Platform
• Have a Scoring system in place, based on the Technographic data accuracy
• Use multiple advanced scrapers, providers, and AI to confirm the presence of the Platform
• Update the CRM Hubspot with contacts data, as well as Task for managing the SDRs activities

How did we proceed?

1. We applied a variation of our proven 5 step TAM Process

• TAM Analysis: Analysis of your TAM, ICP and Channel
• Highly Qualified Database Creation: Build your segmented, highly qualified lead list using Clay and AI Agents
• The Atlas Infrastructure: Create the necessary email and Tech Infrastructure to avoid spam
• The Aries Funnel: Find the Funnel and Message that brings you the highest response rate
• Scaling the Machine: Let’s scale your results by automating, connecting all the dots, and reverse engineering the system

How did we proceed then?

1. Companies Built Up

We started by working on volumes, sending all the companies we scraped to a Clay.com table.

We merged together all companies coming from different Databases, filtered by technographic usage:

• Builtwith
• Apollo
• Sales Navigator

In addition, we added:

• Companies following the customer support platform on its LinkedIn page
• Companies hiring for the Platform roles

2. Companies’ Filtering Process

On each single Database previously collected, we applied a first filtering based on:

• Geographical area
• Employees

3. Companies Qualification Process

All companies that pass the first filtering process are added to a Main Database. Here, each company that is potentially a platform user goes through a waterfall process to verify and confirm the presence of the technographic data.

In order it’s applied:

1. Platform subdomain existence verification, through JavaScript and Zenrows scraper
2. Platform URL creation and scraping process through Google, to find out if existent or not
3. Clay AI Agent to perform an autonomous search

If one of the 3 steps returns a positive result, the company is classified as a Platform User.

A scoring applies to score companies that returned more than 1 valid criteria.

Behind the curtain – Process visual

Behind the curtain – Process visual

4. Contacts’ Finding (Phase 1)

Once we knew what companies used the Customer Support Platform, we looked for the decision-makers at those companies.

To find the decision makers, we applied a double filtering:
First, finding decision makers on 2 Databases: Apollo and Sales Navigator

• We imported all of them in Clay To ensure optimal coverage
• We excluded all duplicate contacts

Then an exclusion and tiering process:

Creating 3 different Tiers, where Tier 1 had the main ICP, and Tier 2 and Tier 3 had secondary decision-makers

Excluded keywords that were out of scope (intern, success, sales, ..)

5. Contacts’ Finding (Phase 2)

For all the companies for which we could not find any contact attached, we applied a second procedure of contacts’ finding.

1. From the main Database, we intersected the information with the contacts’ table to see when a relevant contact was found for the company, and when not.
2. If it was found, no further action was taken
3. If it was not found, a successive contact waterfall procedure followed:
   • With Apollo and specific keywords, built-in Clay
   • If no contact was found in the first step, a following Clay AI Agent (Claygent) looked for specific roles and contact information within the company

6. Email Finding and Validation

For all the companies that passed the test and were not “Unqualified”, we found and enriched their work email. For the contacts in Tier 1 and 2, we searched for the phone number too.

7. CRM Enrichment and Task Generation

All contacts were in the end saved in the CRM Hubspot, passing through Lemlist.
Tasks to call the contact were generated through Make.com.

‍

In Summary over the first 4 months we have:

• Generated an extremely targeted Database, that the client was trying to build for years
• Qualified and Disqualified companies and contacts based on strict ICP criteria
• Saved hours and hours of the client time in sourcing these leads

The Results Achieved – In 4 Months

+6287 In-Target Database

+14046 Qualified Contacts

‍