How to Tell a Klaviyo Operator From a Klaviyo User
The hardest thing about hiring for email is that the surface layer of the job is easy to fake. Anyone can learn the Klaviyo interface in a weekend. Anyone can build a campaign, drop in a product block, and screenshot something that looks clean in a portfolio. The parts that actually determine whether the program makes money are invisible in a screenshot, which means the standard interview, where a candidate walks you through work that looks nice, tells you almost nothing.
What separates an operator from a user is architecture. So ask about architecture. Have the candidate describe, out loud, the full flow structure they would build for a brand like yours, and pay attention to whether they mention suppression at all. A person who has only built individual flows will list them: welcome, browse abandon, cart abandon, post-purchase, winback. A person who has run a program will immediately start talking about how those flows avoid each other, because a customer who is simultaneously in a browse abandonment sequence, a cart sequence, and a replenishment reminder is a customer receiving three emails in a morning from a brand that appears to have no idea who they are. Ask them how a purchase event exits someone from every other active sequence. Ask what happens when someone qualifies for two flows within an hour. The answers separate people fast.
Give them a messier problem next. Describe a list that has been blasted with the same daily promotional email for two years, and ask what they would do first. Weak answers jump to creative: new templates, better subject lines, a fresh design system. Strong answers start with the list itself, because two years of blasting has produced a file where a large share of addresses have not opened anything in eighteen months, and continuing to mail them is actively suppressing inbox placement for the people who do want to hear from you. You want to hear about engagement segmentation, a gradual sunset, and a deliberate warming approach rather than an immediate volume cut that trips filters in the opposite direction. You want them to say they would look at the data before promising a result.
Then ask the diagnostic question, the one that reveals whether they can think causally. How would you tell a deliverability problem from a creative problem? Both look identical at the top level, since revenue falls and open rates sag in either case. The useful answer walks through evidence: performance split by mailbox provider, since a Gmail-specific collapse points at placement rather than copy, delivery and bounce patterns over time, whether the drop is sudden or gradual, whether transactional messages are affected too, whether the decline tracks a specific send or a specific list segment. Someone who has actually debugged this will get specific without being prompted. Someone who has not will talk about testing subject lines.
Ask which single number they would put at the top of a monthly report and why. The reasoning matters more than the metric. If they choose something that improves when you mail fewer people, ask how they would prevent the program from optimizing itself into irrelevance. If they choose revenue, ask how they attribute it and what window they use, then listen for whether they understand that an attribution window is a choice with consequences rather than a default setting.
Finally, ask about a program that did not work. This is the question that most reliably ends the performance. What did they expect, what happened instead, what did they change, and what did they learn that changed how they build now. Anyone who has run email for real has a story here. Anyone who does not have one has either never owned an outcome or is not being straight with you, and the deeper hiring tradeoffs covered in hiring a Klaviyo expert matter far less than whether the person in front of you can answer this honestly.
The portfolio deserves the same skepticism. Revenue claims with no baseline are meaningless, because a program going from nothing to something is a different achievement than a program going from good to better. Be careful with anyone taking credit for a brand’s growth during a period when paid media spend tripled, since email revenue rises mechanically when more people arrive. And treat “we increased open rates by forty percent” as a warning rather than a headline, given that open tracking has been unreliable since privacy protections started prefetching images, and anyone presenting it as their main result is telling you which metrics they optimize for.
Whoever you hire, in-house or through an agency running your Klaviyo email marketing, the test is the same. Can they explain why something worked, in mechanism, without reaching for a screenshot?
