Everyone Agrees the Welcome Email Wins. The Numbers Disagree on By How Much
Pretty much every ecommerce benchmark report out there lands on the same conclusion: the automated email that fires the moment someone signs up beats an ordinary campaign, hands down. Open rate. Click rate. Revenue. Doesn't matter which source you check. Klaviyo's February 2026 benchmarks, built from more than 205,000 brands, put flow-based emails at roughly 3 times the click rate and 13 times the placed-order rate of ordinary campaigns. Omnisend's May 2026 benchmarks zoom in on welcome emails specifically, across more than 27,000 brands: a 35.53% open rate against 30.41% for campaigns, revenue per email of $6.16 against $0.155.
The exact multiples move around depending on whose panel you happen to be reading, and here's a wrinkle worth knowing before anyone cites one like scripture: even Klaviyo's own two public benchmark pages can't agree with each other, not on sample size, not on which multiple gets top billing. So treat any single figure here as one vendor's snapshot, not some fixed constant. And there's one number going around that deserves to be put down for good: an 83.63% open rate for welcome emails, credited to GetResponse's benchmark report, published in 2024 but built on data from 2023, three years old relative to 2026. Real number. Real study. Just not a current one anymore.
Here's what holds up, no matter whose multiple you go with: the window right after someone signs up is the single highest-attention moment a brand is ever going to get with that contact. Every source agrees on that direction, even while disagreeing wildly on the size, and that kind of agreement across sources is itself stronger evidence than any one vendor's own talking point. Weaker, and this is where a lot of marketing writing gets sloppy, are the decay curves that keep making the rounds, all those specific claims about how fast attention fades and by exactly how much. Most of that traces back to practitioner opinion repeated so many times it starts sounding like measurement, not to any one named, dated study. Confidence isn't evidence. So take the shape of the effect as real and leave the exact slope alone.
The Question Nobody Has Actually Answered
A popup. An exit-intent overlay. A product page form. A back-in-stock alert. A footer newsletter box. A post-purchase signup prompt. Six pretty distinct moments, six distinct states of mind, and it only stands to reason that someone who deliberately asked to hear about a restock behaves differently, over time, than someone who finally closed a popup out of pure fatigue after dodging a few of them already. You'd think somebody would've checked this by now.
Nobody has. That specific comparison just doesn't seem to exist anywhere in public form, and that's the part worth pausing on. Plenty of ESP benchmarks and case studies slice their numbers by flow type, welcome versus abandoned cart versus back-in-stock, or by how complex a form is. None of them slice open rate, click rate, purchase rate, average order value, time to first order, or unsubscribe rate by original signup context, for one shared population. That's a genuine hole in what actually gets measured and published, not some polite hedge: nobody publishes signup-source-to-downstream-behavior data.
One partial exception exists, and it's instructive precisely because of how it closes the gap. A signup quiz doubles as its own data-capture mechanism, so a brand running one already knows more about a new contact than any plain form ever would. Revenuehunt's 2026 benchmark report, covering more than 45 million quiz responses across 20,000-plus stores, found quiz completers converting at roughly 2.75 times a typical baseline rate, with quiz-attributed orders running 11 to 15% higher on average order value across most of the stores studied, though the report itself flags that this isn't a clean apples-to-apples comparison, since quiz-takers self-select as more engaged to begin with. It exists because the quiz format collapses "how someone signed up" and "what we learned about them at signup" into one single field, something none of the popup or form-based approaches manage on their own.
What Klaviyo, Omnisend, and the Rest Actually Let You Do
Every ESP looked at here, Klaviyo, Omnisend, Attentive, Shopify, Mailchimp, can natively tag or route a contact by which signup form they used. Solved, and cheap, on every one of these platforms, so let's not pretend that part's impressive. What none of them do natively is keep that signal reliable over time, or tie it to whatever the person was actually looking at the moment they signed up. That's the part that separates a real system from a checkbox.
Klaviyo's $source Is a Snapshot, Not a Record
Klaviyo automatically writes a $source value onto every profile the moment someone signs up, naming either the form or a numeric code, and that value can drive a segment or branch an entire flow. But Klaviyo's own documentation says the value gets overwritten whenever a profile takes an action that sets it again, including filling out any other form, so it isn't a permanent first-touch field. Someone signs up through a product page popup, then later fills out some completely unrelated form, and their original signup source quietly disappears, replaced. Preserving the real first-touch value takes a deliberate extra step: a flow built to copy $source into its own separate custom property, exactly once, and never touch it again. Exactly the kind of infrastructure work no form checkbox hands a brand for free.
Omnisend Tags Every Form the Same Way by Default
Omnisend applies the same generic source:form tag to every single signup form, straight out of the box, no exceptions. Telling a homepage popup apart from a product page popup takes someone going into Settings and manually assigning a separate tag to each form, one at a time, exactly the kind of chore that gets put off forever. Nothing forces that step. Most brands skip it. So the default result, even on a platform genuinely capable of doing better, ends up being that every new contact lands in the identical welcome flow, regardless of where they actually signed up.
The Rest of the Field: Real Primitives, Still Manual
Attentive lets a brand tag a contact by which sign-up unit they came through, and its text-to-join keywords stand out as a genuinely strong exception here, credit where it's due: texting a specific keyword is already, just by nature, a distinguishable, campaign-specific signal, none of the manual tagging discipline the rest depend on. Shopify's Forms app auto-tags by form or page and can kick off a Shopify Flow automation from that tag. Mailchimp handles it through a dedicated landing page. Capturing which form someone used, in every one of these cases, is a solved problem, and the easy half. Connecting that signal to what the person was looking at, or keeping it accurate once the same contact interacts with the brand again, is a whole different problem. It means wiring that signup property to a second native primitive most brands already have running: a behavioral event stream already feeding cart and browse-abandonment flows, reused for welcome branching instead of built from scratch. None of it needs a dedicated customer data platform. It needs someone to connect pieces that are already sitting right there.
The Discount Isn't a Failure of Imagination. It's the Default
If branching a welcome series by signup source takes this much deliberate wiring, why do so many welcome flows land on the same move anyway: a percentage off the first order? Because two of the platforms above ship that discount as the literal, out-of-the-box default, and it's kind of hard to fault a busy brand for taking the path the software hands them.
Shopify Email's own ready-made welcome automation templates go by the names "Welcome email series (discount with reminder)" and "Welcome email series (brand story with discount)." Both either open with or close on a discount code, and neither one branches its message by where on the site someone actually signed up. Attentive's own best-practice documentation goes even further, recommending a discount in the welcome message specifically "to drive immediate sales." None of this is some workaround a brand invented to dodge the harder work described above; it's the vendors' own shipped playbook, which is more damning than it sounds. The blanket discount is the path two of the biggest platforms pave by default. Branching a welcome series by intent is the path a brand has to carve out on top of that, uphill, alone.
Does the Discount Actually Pay Off? Two Studies, Two Different Questions
Michael Lewis's 2006 study, published in the Journal of Marketing Research, drew on customer-level data from a newspaper and an online grocer and found that steeper acquisition discounts were negatively linked to both repeat-buying rate and long-term customer value, across both datasets. The often-repeated headline figure, that a 35% acquisition discount yields customers worth roughly half the long-term value of customers acquired without one, gets carried through multiple sources citing the paper rather than the paper itself, which returned an access error during this research. The study's existence, venue, and direction of finding all hold up fine. That specific number is once removed though, which matters, but isn't a reason to throw the whole study out.
Then there's a 2026 field experiment in the Journal of Retailing and Consumer Services, run on almost 130,000 customers in Japan, which found something that looks at first like the opposite result: redeeming a coupon cut churn probability by roughly 10 percentage points and boosted the promoted product's lifetime value by more than 30%.
Here's the thing though, and it'd be lazy to skip past it: these two studies aren't in conflict. They're answering different questions. Lewis measures what kind of customer pool you end up with when the acquisition itself runs on a discount. The 2026 study measures redemption effects inside a relationship that's already ongoing, closer to a retention or win-back promotion than to a signup-moment discount. Which of the two a welcome discount resembles more isn't settled by either paper, and the honest answer is that it stays unresolved.
One alternative stands out as genuinely better evidenced: the same quiz format that closes the intent-signal gap above also doubles as a welcome-offer alternative to a plain percentage off, backed by the comparative data already cited. Free shipping, a dollar-amount threshold discount, a free gift with purchase, and early access all turn up repeatedly as options brands reach for instead, but no independent, comparative performance data turned up for any of them measured against a percentage discount.
The Honest Takeaway
The signup-source-to-behavior question is a real, documented gap in what anyone actually publishes, not a question that's already been answered and just left uncited. The mechanism to close it, tag a form, keep that tag reliable, pair it with the behavioral data most platforms already collect for other flows, exists natively across every major ESP checked here. Building it is deliberate integration work, not some new category of tool waiting to be invented. Whether that welcome flow should end on a discount, a quiz result, or something else entirely is a question the public evidence still only half-answers.
