Open rate went up. Revenue stayed flat. Sound familiar? Email analytics get cargo-culted harder than almost any other part of ecommerce marketing - everyone watches the same two numbers, draws the wrong conclusions, and makes the wrong changes. Here's how to read what your data is actually saying.
The problem isn't that DTC brands ignore their email analytics. Most founders check their stats religiously. The problem is they're watching a highlight reel when they should be watching film. Open rates, click rates - those are fine as pulse checks. But they don't tell you why performance looks the way it does, or what to do next. If you're making send decisions based on those two numbers alone, you're flying mostly blind.
Click rate tells you who showed interest. Revenue per recipient tells you whether your email program is actually earning its keep. It's calculated simply - total email-attributed revenue divided by the number of people the email was delivered to. It collapses deliverability, relevance, conversion, and average order value into one number that's hard to spin.
When you run a broad promotional send to your whole list and see a decent click rate, revenue per recipient may tell a different story - because a lot of those clicks don't convert, and you spent send volume on people who were never going to buy. Compare that to a tightly segmented send to engaged recent buyers. The click rate might be lower. Revenue per recipient tends to be more favorable - though results will vary by list, offer, and segment quality. That's the signal worth chasing.
Since Apple Mail Privacy Protection started prefetching opens, a chunk of the opens you see in Klaviyo are machine-triggered, not human. This doesn't mean open rate is worthless - it means you have to use it differently. It's still useful for spotting big swings: a sudden drop can flag a deliverability problem or a subject line that genuinely missed. But optimizing subject lines purely for opens, without watching what happens downstream, is how you train yourself to write clickbait that doesn't convert.
Use open rate as a directional indicator. Pair it with click-to-open rate (clicks divided by opens) to understand whether the people who did open actually engaged with the content. A high open rate with a very low CTOR usually means the subject line overpromised and the email underdelivered.
Campaigns and flows serve different purposes and should be judged differently. Campaigns - your newsletters, promos, seasonal sends - are interruptions. You're reaching out on your schedule. Flows are responses to behavior. Someone abandoned a cart, made a purchase, browsed a category. The email showed up because they did something first. That context matters enormously when you're interpreting numbers.
Flows often outperform campaigns on a per-email basis, because the trigger creates relevance - though this varies by program. Comparing your post-purchase flow to your weekly campaign is like comparing a repeat reorder to a first-time browse: the intent level isn't the same, and neither are the conditions. Benchmark flows against themselves over time, and campaigns against other campaigns. Keep the two buckets separate.
A lot of brands look at unsubscribe rate, see a small number, and move on. That's understandable - the absolute figures do tend to be small. But the trend matters more than the absolute level. A steadily climbing unsubscribe rate across your campaigns is telling you something your click rate might be hiding: you're sending content that people opted out of wanting. It often shows up before deliverability problems do.
Look at unsubscribe rate by segment, not just by send. If it's consistently higher among people who bought once six months ago and haven't purchased since, that's a segmentation problem - you're sending engagement content to people who've already mentally checked out. The fix isn't a better subject line. It's a different message, or a winback sequence, or pulling them out of your main list entirely. A simple rule to work from: anyone who purchased once, hasn't opened in 90 days, and hasn't bought again should move out of your main campaign list and into a short winback sequence - say three emails - before suppression.
Weekly or post-send stat checks are useful for catching fires. But reactive monitoring picks up noise as often as signal. A quarterly audit gives you enough data to spot genuine trends - a drift in deliverability, a segment quietly churning, a flow that's gone stale - without waiting so long that problems compound into something harder to reverse. Build the habit around these questions:
Klaviyo's default attribution window is generous. If someone clicks an email and buys five days later, that revenue gets credited to the email. That may or may not reflect how the purchase actually happened - they might have come back through paid social, a Google search, or a direct visit. This isn't a reason to distrust email attribution entirely. It's a reason to pick a consistent window (many brands prefer a shorter click window, like one or two days) and apply it consistently so you're comparing apples to apples over time. Worth noting: a tighter window undercounts email's influence on considered purchases; a looser one overstates it. Neither is wrong, but you have to know which bias you're living with.
The goal isn't perfect attribution - that doesn't exist anywhere in marketing. The goal is a consistent internal benchmark that lets you see whether your email program is trending in the right direction. Change the window and your historical comparisons break. Pick one, document it, leave it alone.
Most analytics problems in email trace back to the same root cause: the flows and campaigns weren't built with measurement in mind. When Kaydence reads your store and generates Klaviyo flows - brand-voice copy, on-brand creative, audience segments - each flow is structured around specific lifecycle triggers that make it easier to evaluate performance in isolation. For example, a post-purchase flow built with Kaydence separates first-time buyers from repeat purchasers at the trigger level, so you're benchmarking each cohort independently from day one. You import the flows you want into your own Klaviyo account, review and revise everything before anything goes live, and then you own the sending. What you get is a starting point that's already organized the way good analytics require: segmented, triggered, and separated by purpose. The DIY tool is in private beta (public launch Q3 2026), but the done-for-you service is available now if you'd rather move faster.
Better data habits don't require a fancier tool. They require asking better questions of the data you already have. Start there.
Most DTC brands segment by purchase count and call it a day. Here's how to slice your buyer list by what people actually bought - and why it rewrites what you say to them next.
Most DTC broadcast emails get ignored because they lead with the offer. Here's how to structure promotional sends so readers engage before they see the ask.
Kaydence is the done-for-you Klaviyo email team - we build, write, and run your flows, campaigns, and creative. Book a free teardown and we’ll show you what we’d ship first.
