Crafting Jobs to Be Done: A Faster Path to Branded Email Creation in Klaviyo
Before Klaviyo built a brand management feature, I asked what job users actually needed it to do, and found they’d already hired Canva for it.
Overview:
The ask was to evaluate the need for a feature set in the brand management space, one that had already been scoped out but required further engineering and business investment before moving forward.
We knew from customer feedback via Support and Customer Success that one of the most painful tasks for users was managing their brand and formatting brand assets within their email communications.
Project Scope
Company; Klaviyo
Timeframe: 8 weeks
My Role: UX Researcher
Teams Involved: Design, Engineering, Product Management, Data Science
Methods: Jobs-to-be-Done, Survey
Tools: Figma, Zoom, Qualtrics, Slack
Constraints:
Technical: Building tools to support branded content required significant engineering effort, so our recommendations had to be prioritized by impact against effort.
Participant access: Reaching the right users took coordination. We had to work around customer communication schedules and avoid over-contacting the same accounts.
Product Area: Content
The goal of brand management tools was to give users one place to manage their logo, headers, footers, and contact information, so every email, landing page, and form had a consistent look and feel.
Key Question:
What "job" do users expect Klaviyo to do when it comes to managing their visual brand?
Business Goal:
Reduce time to value in building branded content within Klaviyo.
User Goal:
Better support users in creating branded emails inside Klaviyo.
Research Phases and Key Insights
Phase 1: Jobs to be Done Interviews
Goal: Surface user goals around managing a visual brand. These interviews revealed the underlying "jobs" users were trying to accomplish, along with the workarounds people relied on in the absence of a purpose-built tool.
n = 11
Key Takeaways:
The majority of participants employed Canva to manage their visual brand, telling us that it was easy to use and build their branded content there while using Klaviyo to send the emails.
One of the principal challenges identified in the research was to connect Canva and Klaviyo to enable users to send the branded content build in Canva through Klaviyo.
Phase 2: Survey
Goal: Quantify which external platforms users were already relying on for branded content, and identify which specific brand elements they were designing outside of Klaviyo.
n = 73
Key Takeaways:
Canva usage was widespread among users for creating branded content.
Klaviyo was perceived primarily as a tool for sending and tracking content, and for building simple, non-branded emails, rather than a platform for branded design work.
From the interviews and the survey, the data indicated that the proposed solution we were evaluating duplicated tools people already had and were working well.
Deliverables
Research findings decks with recommendations
Defined job statements
Cross-functional alignment workshop to determine next steps
Outcomes Tied Back to the Challenge
Redirected investment away from duplicate tooling: Showed that a native brand-building feature would duplicate Canva, which users already relied on and liked. That saved engineering, design, and budget for a feature users weren’t likely to adopt.
Prioritized a Canva integration instead: Made the case for connecting Canva and Klaviyo, so users could keep designing where they already worked and send through Klaviyo.
Defined jobs to be done: Created job statements the team could design against, grounded in how users actually split their work: designing in Canva, sending and tracking in Klaviyo.
Laid the groundwork for future work: Created an early customer journey map that later projects built on.
What I’d do differently: Better connect findings to shifting business goals. Priorities changed during the project, and I’d revisit how our findings mapped to each new goal as it shifted. That way the research would stay clearly relevant to the decisions being made at each point, not just the question we started with.