How DigitalOcean built custom messaging for 50,000 contacts within a week of deployment

Andrew Park on October 9th 2018

Zoom with margin

This post is a guest submission by one of our partners at Tray.io for Segment’s Chain Letter blog series. The Chain Letter series profiles clever uses of Segment Connections partner tools that, when chained together, lead to some pretty advanced programmatic models, custom messaging strategies, and more. Thanks to Tray.io for sharing DigitalOcean’s story!

—Segment

The marketing team at DigitalOcean knew it couldn’t usher in growth for its more than half a million global customers using traditional demand gen tactics. David Dorman, Director of Growth and  Demand Gen, and Andy Hattemer, Senior Growth Marketing Manager, needed a system that would deliver highly customized, context-heavy outbound and nurture messaging to individual users at different points in their customer journey. Creating and executing on an individualized messaging system at scale involved several hurdles. This is the story of how DigitalOcean evolved a reactive marketing approach, previously powered only with Salesforce and Marketo, into a highly personalized messaging strategy by layering in Tray.io, Clearbit, and Segment.  

A two-system state of the world

While DigitalOcean was using Salesforce for customer relationship management and Marketo as its way to integrate customer messaging into Salesforce, the growth marketing team quickly realized that these two solutions, while popular for certain marketing automation and CRM functions, weren’t going to scale without additional functionalities.

The marketing platforms lacked the customized multi-directional, multi-app syncs that DigitalOcean needed - the kind of syncs that would flow important customer data to and from Segment to both enrich user data in other applications, as well as to update Segment records afterward to ensure DigitalOcean’s campaigns were always properly targeted. The team also found itself up against Marketo’s low API call limits - typical for most marketing platforms, but a bottleneck for a team that needed to route huge amounts of mission-critical customer data. The solution was to enrich their customer data and craft messages that spoke to the distinct needs of each user while being able to deploy these messages multiple times per day, at scale.

Adding in a customer marketing data stream

As a Tray.io and Segment customer, DigitalOcean realized that it could use Tray.io’s General Automation Platform and Segment’s Personas product to develop Audiences to send to their downstream tools like Marketo to tailor campaigns for each user. DigitalOcean created a system that chained together:

  • Segment Connections and Personas to inform the messaging customization and enrichment process. Once the enrichment and customer profiling process was complete, DigitalOcean used Segment Connections to build email metadata to properly identify which users would be appropriate for which messaging

  • Clearbit to enrich each customer with full lead information, including name, email, and location

  • Salesforce to confirm DigitalOcean customers against their account records in the CRM

  • Tray.io to integrate all of DigitalOcean’s cloud-based services, creating a sophisticated automated workflow that would enrich leads with initial webhook data, Segment Personas data, and Clearbit, compare against Salesforce account info, and prep Segment email metadata

Step 1: Enrich Messaging, Add Personas, Enrich Lead Data

Zoom with margin

DigitalOcean built a complex workflow on the Tray Platform that automatically fired off multiple times per day according to a scheduled trigger. The first part of the workflow, once triggered, enriches the current messaging queue via webhook for its many messaging parameters (_including _Attributes_Timestamp, SenderID, MessageAttributes_City, MessageAttributes_PostalCode), then sends a GET webhook request to pull Segment Personas details from a records.segment.com URL.

Next, the workflow combined this persona information with enrichment from Clearbit using a Clearbit Output function for specific lead details (including Person_FullName, Person_Employment -Domain/Name/Title/Role/Seniority__, and Person_EmailProvider__, Geo - City/State/Country Person_Company_Metrics - NumEmployees/MarketCap/AmountRaised/AnnualRevenue) to give the clearest possible picture of the target customer.

By knowing where each company is headquartered, DigitalOcean could ensure its messaging would be sent in the appropriate language, and the most relevant users could join their local Meetups. By connecting Crunchbase data, the team could ensure it was presenting appropriate messaging based on their users’ company size (for instance, marketing messaging for a small-to-medium-sized business won’t resonate appropriately with an enterprise customer).

As a result of this series of steps, DigitalOcean’s team would ensure it had enriched profiles for each of its thousands of customers in its messaging queue, profiled with Segment Personas and enriched with Clearbit data as well as with employee size details sourced from Crunchbase to ensure it was sending messaging appropriate to specific job titles and company sizes. For instance, a specific marketing message might hit home with an Associate IT Manager at a small-to-midsize company; but that same message would be unlikely to resonate with a VP of Engineering at an enterprise firm.

Step 2: Tray Helpers, Confirm CRM Account Details, Email Metadata

Zoom with margin

Once the automated workflow in the Tray Platform had enriched and profiled the company’s customers, DigitalOcean then directed it to use a variety of customized Tray Platform helpers, such as running an API query at api.crunchbase.com to confirm whether a customer’s company with missing data had a Crunchbase listing, then loading in a new profile if that company’s Crunchbase profile had not yet been added:

The workflow then made a call to Salesforce to Lookup Salesforce ID Output for Accounts and Account ID to confirm whether the customer’s company was already listed in the CRM. Finally, the workflow made a Segment Email Metadata Output call to identify users matching the enriched profile to prepare the appropriate customized messaging stream for those users.

Hundreds of thousands of custom messages sent monthly

As a result of using Segment’s Personas functionality and its ability to interact with email metadata, DigitalOcean quickly scaled from traditional marketing automation to enrich and prepare custom messaging for some 50,000 contacts in its first week of deployment. DigitalOcean is now sending out personalized messages at scale to their prospects and customers with the right message at the right time. This gives DigitalOcean the speed, visibility, and costs savings to be proactive with their marketing across all customer touch points.

The State of Personalization 2021

Personalized, 1-to-1 marketing is table stakes for today's digital-first businesses. But as consumers' expectations rise, are they equipped to deliver these experiences?

Become a data expert.

Get the latest articles on all things data, product, and growth delivered straight to your inbox.