Your Email Tests Aren't Statistical (And Here's Why)

You really do have to focus on nailing positioning and the structure first. Don't test multiple variables at once. Just isolate one change per test.

You know, what's smart about things like this is that you're basically building statistical significance through repetition.

Basically focus on building like this creative library to compound your learnings. Punishment emails, getting the timing right there really can have significant impact on the conversion rate.

Incremental testing. So if brands that are testing consistently over 6 to 12 months, They massively outperform brands that test occasionally.

I bet use that kind of HCE framework that we talked about, which is the hypothesis, control, and evaluate.

Bad tests are worse than no tests, so test with discipline.

From the open to the click, they're the masters of the flow. Just watch your list grow and your leads turn to go. Did we retain you so far, or do we have to extend it? If you want results, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it.

What is up everyone? Welcome back to Send It where we discuss anything and everything retention marketing specifically in email and SMS marketing and e-commerce. My name is Jimmy Kim and I'm here with my co-host Chase Diamond. Chase, today we're talking about something that's gotten way harder in the past few years.

Yeah, what is it?

Well, we're talking about testing.

Yeah, testing. I mean, I hear this from brands often, right? We wanna test more, but we don't know if we're getting any real signals anymore.

Yeah, and I think it's a lot to do with the old playbooks that don't work anymore. You can't just A/B test 2 subject lines looking at open rate and call it a day anymore, right?

The good old days, right? But yeah, now it's all about smaller segments because of Apple's mail privacy protection, which leads to a lot of noisy attribution. All of that has made traditional testing so much harder.

Yeah. So today on this pod, I wanted to kind of walk us through how to run creative testing for email marketers in 2026. And not just a bunch of theory from 2018, but really frameworks that are working right now.

Let's do it, man. And more importantly, we're going to talk about how to avoid running tests that give you, you know, misleading conclusions.

Yeah, because bad tests are worse than no tests. And at least with no tests, you're not making any decisions based on false data.

Exactly. So without further ado, let's dive in.

All right, Chase. All right, so let's get right into it. So let's talk about diagnosis. Why do you think, as you've seen from your side, that A/B testing feels much harder than it did in the last few years?

I think there's kind of 2 big structural changes that happened that, you know, broke kind of the old testing playbook.

Okay, let's go through those. What's the first change you think about?

Yeah, the first one is small sample sizes. You know, they're shrinking. I think this though is actually a good problem to have, but does make testing harder.

What do you mean by that? Why are small sample sizes shrinking?

Yeah, I think it's because marketers are segmenting more aggressively now, and as they should be, for 2 reasons, right? One, deliverability, which I know is your specialty. Gmail and Yahoo both require better and more engagement. So you can't blast entire lists anymore. And then alongside that, number 2 is personalization. You have to have More relevant messages. Those are gonna convert better. So brands are starting to send to smaller, tighter segments.

Yeah. So instead of sending to like 100,000 people, you're sending to maybe like 25,000 people now.

Yeah, that's right. Which means when you split that into A/B tests, right, if you're doing 10% like you did before, you used to get 10,000 for each variation. Now you might only have 2,500 per variant, right?

I mean, if you're testing more than 2 variants, it kind of And when you say that, what it means is smaller sample sizes means it's harder to reach statistical significance, really.

Yeah, that's right. You essentially need more data to prove that one version's actually better and not just sheer luck.

Yeah, I get that one. Okay. What about the second change you're thinking about?

Yeah, that one is related to the Apple Mail privacy protection. So that one launched, I don't know, 5-ish, 6 years ago, whatever, sometime around 2021. And that completely destroyed the open rate as a reliable metric, unfortunately.

Okay, so for people listening that have no idea or have been living under a rock, could you tell them more about what that is?

Yeah, sure. So Apple essentially started preloading images in emails before a user actually even sees them and has a chance to open them. So your ESP, right, like the Omnisend, the Klavios of the world, et cetera, they think someone actually opened the email when it really was just Apple servers downloading the image.

So yeah, I mean, open rates really got inflated over the time.

Yeah, big time, dude. You know, if 40 to 60% of your list uses Apple Mail, your open rates could be 10, maybe 15% points higher than reality.

Yeah, and that ultimately means that you can't trust these open rates anymore for subject line testing anymore.

Exactly right. And essentially, like, the old playbook used to be like, test subject line A versus subject line B. You look at the open rates and the winner essentially gets selected based off the higher open rate. That does not work anymore.

Yeah. So what does that mean practically in, in your eyes?

So it means testing has to move deeper into more of like behavioral type metrics. So you, you can't stop at opens. You actually have to go way Beneath the surface. So things like click rates, did people actually engage? Site visits, did they go to your website? Things like added to cart rate, did they show purchase intent? Conversion rate, did they buy? And then revenue per recipient. So how much did each variant actually generate in terms of revenue?

Yeah. And these are actually business outcomes instead of vanity metrics now, huh?

Yeah, that's correct. And I think that's This is actually better for making decisions. But it does require a lot more discipline than most folks are used to.

Okay, so I like that. Let's talk about the mistakes that we're seeing first. Let's talk about what the biggest mistakes you see brands are making when they test. What are you seeing?

Hey, it's Chase. Got 30 quick seconds? We built something specifically for you. Jimmy and I together have over 20 years of combined experience in email marketing, and we finally put everything together in one place. And it's called Ecom Email Certified. I know it's fancy, but it's gonna be great whether you're brand new or running campaigns for 7, 8, 9-figure brands. It covers it all. Deliverability, AI, copywriting, design, strategy, SMS, literally everything. Has over 175 total modules totaling 25+ hours of content. Even has quizzes, a community, and you're gonna get a real certification. It's way better than college, right? Don't tell your professor that, but it is. And we've designed it to take you from beginner to Certified expert. We're talking about the kind of marketer who actually moves the needle and drives real revenue. Check it out at ecomemailcertified.com. And now back to the pod. Yeah, I'd say testing too many variables at once is still something that people fall into. I think this is a really common mistake and it just ruins the test.

Yeah. So like the everything mistake, right? So what, what do you see brands do? Give us some examples of what they do.

Yeah. So for a little bit of an extreme example, but they'll change the subject line, the preview text, the email layout, the CTA copy, the imagery, the offer, all at the same time in the same email, dude.

And all they do is look at which one of those versions wins, right?

Yeah, that's right. And let's say version B performs 20% better. Woohoo. That's great. It performed better, but now you have no clue why.

Because they changed 6 things.

Right. You know, was it the subject line? Was it the image, the call to action, the offer? Like what moved the needle? You don't know. I don't know.

So even though they have a winner, you can't just apply all the learnings to a future campaign because you don't really have learnings.

That's right. And worse, you might kind of attribute it to the wrong variable that caused the lift. You might be confident that it was the subject line, but that might not have been the thing that moved the needle.

So what's your golden rule? What's the right way to test?

Dude, singular focus, one test, one hypothesis. You have to isolate a variable and then test that, you know, by itself.

Alright, let's dig into that. Give, give me an example for the audience.

Yeah, sure. So let's say you want to test whether social proof increases clicks or not. Your hypothesis is going to be that adding customer reviews above the fold is going to lead to an increase in click-through rates.

Okay, keep going.

Yeah. So what's going to happen, right, is you'll have 2 versions, version A, will be kind of your typical email that you send with no reviews. And then version B is going to be the exact same email, but all you're going to do is add like a review type quote above the CTA.

And everything else stays the same.

That's correct. Same subject line, same call to action, same layout. The only difference is the addition of the review quote.

And then if version B wins, then you know it was the social proof, right?

Right. And now you have learnings you can apply, right? Social proof above the fold in this particular email or in these multiple emails increased our clicks by X%, right?

Yeah. Okay. But I know what people are thinking. Is there ever a time you think you can test multi-variables at once, like multiple variables?

Look, I want to say no, because the general population isn't going to need this, but the, the, the actual answer is yes. However, it is quite advanced. Um, and it's called multivariate testing. So essentially you're going to need a lot, a lot of traffic to do this properly.

Like how much traffic?

Um, yeah, hard to put like a specific number on it, but you're going to need enough volume to be able to test every single combination of variables and still have a significance that you trust. Right. And unfortunately most brands just don't have this.

Yeah, I agree. I think most brands just stick to one variable at a time.

Yeah, it's way easier, way simpler anyways.

All right. So, okay. If we're testing one thing at a time, what should we actually test?

So I like to think about testing in 4 different layers, um, from basically the biggest impact to the kind of smallest impact. So that first one, layer 1, is positioning tests. And this is about the core story you're telling.

Okay. I want to hear some examples of this one. I like this.

Cool. So, all right, so kind of a couple positioning tests. So this could be something like problem-focused versus an aspiration-focused. So a problem could be something like struggling with dry skin, where kind of the aspiration would be something like, imagine, you know, clean, glowing, you know, moisturized, healthy skin every morning, something like that, right? And then the next could be like product versus lifestyle-focused. So it could be like our moisturizer has, you know, whatever type of acid, which is like, you know, if you know about that ingredient or that type of acid or whatever, maybe that's helpful. Or the other one could be something like, here's your new morning ritual. And then the last one here would be like focusing on education versus promotion. So the first one would be like how to build a skincare routine versus 20% off all skincare.

I like that you jumped right into like big strategic differences here.

Yeah. And these tests often create the biggest lift. It could be anywhere in the ballpark of like 30-ish to 50% differences in conversion if they're strong, right? If you do this properly.

Okay. So wait, why do you think, and I think I know the answer to this, but why do you think positioning tests have such a big impact?

Yeah, I'd say it's because they essentially change like the entire frame of the message. You're not just tweaking copy, you're actually changing like how your customer thinks and perceives about your actual product and company.

Yeah, that makes sense. And you're not going to— and if you're only going to run one test a quarter, this, this should be the positioning test, right?

Yeah, absolutely. And then the next one, right, layer 2 is more so focused on like structural tests. And this is about like how the email's actually built and created.

Okay. More examples here. Tell me how, like what?

Sure. So, uh, you might wanna do something like long form versus short form. So it could be like a, a 500-word story versus like a really tight, concise 100-word message. Um, the next part could be like image first versus copy first. So do you lead with like a hero image at the top? Um, or do you lead with like a headline and copy first? Um, the next one could be like focused on call to actions. Do you have a single call to action? Or do you have multiple? So maybe you have one button or one link versus 3 different types of buttons or links, right? And then the last one, this is kind of one that I feel like is so heavily debated, is like the whole plain text versus like the beautifully designed HTML type emails. So you're gonna test text only versus these designed emails that e-commerce brands just love to do, right?

Yeah, these are solid tests and I would imagine these can have some pretty big impact. What have you seen?

Yeah, they're strong. Whereas if the first one we talked about was like maybe like 30-ish to 50%, this is probably like the structural test is probably something in the ballpark of like 15 to 30-ish. So not as big as positioning, but dude, still quite significant. Hmm.

That's interesting. You got any structural tests you've seen recently that might've surprised you?

Yeah. Um, one that I recall offhand was something for like a wellness brand where it was like short form versus long form. I mean, everyone assumes short would win because people don't read emails, right? We hear that all the time. Yeah.

But I'm guessing the long-form won then, right?

Yep. Correct. And I think it was by about 22-ish percent. Essentially what the longer email did is it told a story that built trust and the short email really just felt more like generic and kind of rushed in this instance.

Yeah. Marketers do this all the time. They make assumptions that can be wrong and that's honestly why you test, right?

Yeah, that's facts. And then the next layer, right, layer 3, is more focused on like creative tests. And I think this is where most people start, but it's one of the smaller kind of impact layers there is.

Let's dig into that. Tell me about some creative tests you've seen.

Yeah, so these are a little bit simpler, right? Like it could be a product photo versus a lifestyle photo, or someone literally could just change the call to action wording, shop now versus get yours. Or people love this one for whatever reason, button colors. I'm going to change this color from green to orange, right? Okay. Congratulations. And then design layout, right? I'm going to design this centered or I'm going to left align this.

Yeah. And when I see these types of tests, they're typically smaller lifts.

That's right. You know, again, it varies, but maybe like 5 to 15-ish percent, which again, it's worth testing. I'm not trying to knock it entirely. But I think you really do have to focus on nailing the kind of positioning and the structure first.

Yeah. So don't start there. This is definitely an optimization point.

Yeah, that's right. Because if your positioning is wrong and a different button color isn't going to save you. And then the last one is like layer 4 is kind of like timing tests, essentially when the message is sent.

Yeah. Send time optimization, right?

Yeah, that's right. Like you send In the morning or at night? Do you send on a Tuesday or a Saturday? Do you send every single day or maybe 2 to 3 times a week? For your replenishment window for like people that buy topicals or consumables, are you sending that at day 25, day 27, day 30? What does that look like?

Yeah, and timing can sometimes make a big impact on certain type of emails. Would you agree?

Yeah, I think for like the replenishment emails, getting the timing right there, Really can have significant impact on the conversion rate.

Yeah, that makes sense. So in the order that we've kind of gone through in the layers, it's positioning first, then structure, then creative, then timing.

Yeah, right? It's like start with the biggest levers first and then work your way down.

Okay, so now let's move on. So we've established that sample sizes are shrinking. How do you test when you don't have massive lists? Hey, it's Jimmy, and I've got 30 seconds to tell you something that will change how you get inspired on your very next marketing campaign. You know when you're screenshotting those emails in your inbox and Slack and pasting them here? You saw it on Twitter, trying to remember what branded that killer Black Friday sequence. Well, guess what? Inbox.ai just solved that. It's an AI-powered database with over a million real emails from the best DTC brands. Ridge, Fenty Skin, Casper, they're all in there. You can search by promotional type, journey stages, whatever you need. Get AI breakdowns of what worked and why, track your competitors, and even request brands that are not in there. No more hunting through screenshots or saved folders. Inbox.ai turns your inspiration into revenue. Check it out at inbox.ai, link in the show description.

I think there's kind of 3 approaches that come to mind when you have these smaller segments. The first approach is sequential testing. So instead of splitting one campaign in half, you're going to run tests across multiple sends over time.

Hmm, that's interesting. Can you explain how that works for everyone?

Sure. So in week 1 of doing this, you might send version A to your entire segment and then you're going to track the results.

Okay.

And then from there, right, next week, so week 2, you're going to send version B to your entire segment and you're going to track the results.

And then you just compare the performance across the 2 weeks.

Yeah, exactly. And here you're kind of using time as your control, um, you know, instead of splitting the list.

Okay. But what are the downsides here? Cause it sounds like there might be.

Yeah, there are. I mean, for every pro, there's a con. I think time-based factors can impact results. If week 2 had a holiday or a major news event, that might skew your data.

Yeah. So you need to control for external factors basically.

Exactly. But if you're consistent with your testing schedule and you track patterns over multiple tests, I do think sequential testing works well for smaller lists. Cool. All right. So let's talk now about this second approach. This is going to be focused on campaign clusters. So instead of testing one email, you essentially test a pattern across, let's say, 5 to 10 different emails.

Ooh, this is something I like doing. Can you explain this for the listeners a little bit?

Sure. So let's say you want to run a test about whether adding urgency language is going to increase conversions or not. So instead of just doing like a simple one-off test, We're going to say, let's do 10 campaigns. So campaigns 1 through 5 have zero urgency language whatsoever. And then campaigns 6 through 10 are going to include urgency language just to see what that does.

And then you just compare the averages of performance against each group again.

Right? Right. And if you, if you notice that like urgency consistently outperforms across like multiple campaigns, you're going to see that it's an actual pattern, not just blind luck.

Yeah. And you know, what's smart about things like this is that you're basically building statistical significance through repetition, basically.

Yeah, exactly. And it, I think it works like great for smaller lists because you're not splitting any single campaign. You're just more focused at like looking at patterns and data over time. And now for the third approach, it's basically what we're going to do is we're going to use leading indicators instead of waiting for revenue.

What do you mean?

So, um, sometimes revenue can lag. It could take 3 to 7 days to show up. If you're testing on a small list, you might not have enough conversions to reach, you know, this data significance.

Yeah, I think I know what you're talking about, but what do you think you should be tracking instead in this case?

So you want to look at kind of like earlier signs that correlate with revenue, things like the click rate. That is something you're going to be able to see immediately. Product page visits. So that's going to be something that should load and see within probably hours. What is your add to cart rate over, you know, 24-hour period? And what's the time on site like? How engaged are your subscribers and customers?

Okay. So if version A has like a 5% click rate and version B has a 3% click rate, you can confidently say that version A will probably drive more revenue, even if you don't have conversion data yet.

Exactly. And these leading indicators, they're going to let you make decisions faster with these smaller datasets that, you know, a lot of us are working with. Hmm.

That makes a lot of sense. Okay. I want to shift focus here and let's talk about something I've noticed with great email teams. Basically, they ultimately all have great creative libraries.

Yes. I think this part's huge, right? They're not starting from scratch on every single send.

Yeah. And what does that mean for testing in your opinion?

So I think like good testing starts with good hypotheses and good hypotheses come from knowing, you know, what's worked in the past.

So you're basically taking historical information or proven patterns from the past and you're iterating on there and you're not guessing in the dark basically, right?

Yes, that's correct. We're essentially like a creative library is basically just like a swipe file of like high-performing campaigns with metrics, you know, strong layouts and templates, you know, proven hooks and subject lines. What call to actions have you had success with? I mean, what type of imagery styles are the most effective for your audience?

And every time you run a test, you're kind of adding the winner to your library.

Yes. So your library is going to get stronger and greater over time.

I like this. I haven't heard about this one. How do you think someone should build this?

3 steps. Number one is just document your winners. So as you see things, as things are working, just screenshot or kind of write a note for every campaign. That's performing above like your average benchmark. Tagging by category. So looking at like post-purchase and then promotional, educational, et cetera, and note what made it work, right? Like you're testing social proof above the fold that led to a 15% increase. So just notate what's happening as it happens so you don't have to just try to remember in the—

That makes sense. I like this. So you're basically building an institutional knowledge Creative folder, basically. Yeah.

And look, you're gonna have different team members involved. People are gonna come and go. You might work with external parties. Having this all as like a common aggregated source is super helpful.

Yeah. And not to put a shameless plug in here, but inbooks.ai is a great way to do this.

Absolutely. Inbooks is basically a searchable creative library of over a million emails. From top brands. So you track your own stuff and then you also can look at the external market.

Yeah, and just to add there, if you wanna test like replenishment emails, you can go to brands that are actually sending them and see what they're sending consistently.

Yeah, and if you wanna test VIP offers, education sequence, abandoned cart recovery, et cetera, you can literally see proven patterns for brands that are crushing it.

Yep, and if you can see those patterns and you get a better hypothesis on what to test, Right, because the creative testing starts with, you know, as much idea generation as possible.

And when you see what's working in the market, you're going to be able to piggyback and generate your own ideas even quicker.

Yeah, so instead of guessing, maybe we should try adding a testimonial, you can say, hey, I saw that 3 of these top brands, my competitors, whoever it might be, using customer quotes above the fold, we should probably test that too.

Yeah, look, I think in today's day and age, it's about shortcutting success. And you do that by testing informed hypotheses. Not just random ideas.

Okay. I like that. All right, let's, let's start talking a little bit about mindset for a minute because, okay, I think the best marketers aren't just testing, they're learning faster than anyone else.

Yeah, I think that is everything. Like the goal isn't to just run like endless tests and blind tests. The goal is to increase your learning velocity.

Ooh, I like that. Tell me about learning velocity the way you think about it.

So I think for learning velocity essentially is like how fast you can generate insights that change your strategy. You know, the faster that you learn what works, the faster you can actually scale it.

Yeah. So you're not running 100 tests basically, you're running this right test and extracting the insights quickly.

Yeah. I mean, dude, I'm such a lazy marketer, right? I'd rather run like 10 high-quality tests with clear learnings than 50 random tests that may or may not teach me much. Yeah.

Okay. So tell me about increasing learning velocity. What would you do?

So number one would be essentially focused on testing bigger variables first. So like the positioning and structure that we talked about a little bit ago, not just button colors. Number 2, I think most of us could do a better job at documenting everything that we're working on and everything that's happening. And after you send an email, go review it. Most people send it, they run a test, but they don't actually review it. So every test should have a hypothesis, results, and some kind of learnings for better or for worse. And you do need to share these learnings across the team. So if one person discovers everything that's interesting and that's working and that's winning, everyone should know because what's working for retention might actually work for CRO acquisition, et cetera, or vice versa. And then build on your wins. So don't just celebrate a winning test and move on. Keep iterating on it.

Ooh, I like that. Build on a win. Give me some examples of building on a win that you've seen. Hey, quick 60-second break to shout out our title sponsor, Omnisend. If you don't know them, Omnisend is an all-in-one email and SMS platform built specifically for e-commerce. They've trusted by over 155,000 merchants with a 4.8 rating star on Shopify's marketplace with over 2,800+ reviews. People love using them. Their support is incredible, responsive team, great help docs, and they actually show up when you need them. 3, look, everyone loves saving a buck. They've got great features and great pricing. Listen, I've heard this inside the community that someone actually dropped their bill by almost 50% and then included their SMS costs, which is absolutely wild. Another agency that we talked to shifted all his clients over when he discovered that he can save them 30 to 60% off of their actual monthly bill. And they're taking that money now, reinvesting into ads, and that real money is going to come back into your pocket too because you're helping them out or whatever it might be. With all this, I literally just helped Chase migrate his personal list to Omnisend not too long ago. It was super easy. They got team, they have an automated solution. It's really great. So if you're looking to scale your email and SMS game, check out omnisend.com/sendit. Tell 'em Chase and Jimmy sent you. Now back to your pod.

Sure. So, um, say you go back to like that example of testing social proof above the fold and maybe it increased clicks by 20-ish percent or so, right? Okay.

That's a win.

Um, don't stop there though, right? Now you need to test like 1 review quote versus 3 review quotes. A text review versus a photo review, a customer review versus some kind of like expert endorsement.

So you're going deeper into what makes that first test work.

Correct. And each layer of testing is going to help compound your learning. So, and with that, right, like, here's the thing about learning velocity, like it's going to compound, you know, the more you learn, The better your future tests become.

Because you're starting at a higher baseline each and every time.

Exactly. You know, if you learn 10 things this quarter, for example, that can improve performance by 5%, you know, in theory, that could be like a 50% improvement compounded.

Yeah, incremental testing. So if brands that are testing consistently over 6 to 12 months, they massively outperform brands that test occasionally, I bet.

100%. And because they're building this advantage, That's knowledge that really is hard to catch up to.

Yeah. Okay. I want to, I want to close with a simple framework because if someone's listening and they want to start testing better today, what do you think about? What's the framework that Chase Diamond recommends?

Yeah, every test should follow 3 steps. Let's call it the HCE framework. So it's hypothesis, control, evaluate. So first is you're going to state your hypothesis clearly, not like, let's try this. Or let's try that, but I believe X, Y, and Z change will equal this result because this reason.

Okay, wait, wait, wait, give me an example a little bit better, like use real words there.

Sure, I believe adding social proof above the fold is going to increase click-through rate by at least 10% because customers need trust signals before they feel comfortable clicking.

You know what I like about that? It's actually a specific prediction with reasoning.

Right. And it forces you to think through like why you're actually testing something. And then second, right, you're gonna control your variables. Only change one thing at a time.

Okay. And you said document everything so it stays the same, right?

Correct. Right. Same subject line, same send time, same segment, same everything except the one variable that you're focused on improving.

Okay.

And I, I'll kind of point out here, a lot of tests fail because they change too many things at once.

Exactly. And I think discipline in this control phase is what makes the tests valid. And the third one would be like evaluating the right metrics. So not opens, right? But looking at like clicks, conversion, and actual revenue.

Okay. Here's a question for you that I thought about. What threshold for declaring a winner? Like what's your threshold?

Yeah, I think statistically you want at least 95% confidence. And some sort of like meaningful difference. And most testing calculators will actually tell you this.

But I think you would also just say use common sense, trust your gut, right?

Yeah, absolutely right. Like if version A gets 100 clicks and version B gets 103 clicks, that's not a real difference, right? Even if it is statistically significant.

Yeah, in the data that's like a 15% difference, but it's really not actually much.

Exactly. And then document the learning and add it to your kind of creative library.

This is good. All right, let's wrap this pod up and do a little recap of what it is, I would say, how to run good testing that actually works in 2026.

Yeah, sure.

Yeah.

So smaller segments and privacy changes, they broke the old testing playbook. I think that's pretty clear. Don't test multiple variables at once, just isolate one change per test. And you wanna test in order of like the most significant potential impact. So that's from positioning to then the structure to the creative and then to the timing. And you're gonna want to test using things like sequential testing, campaign clusters, and leading indicators for those with small lists, right? And then with that, like whatever you do, focus on building like this creative library to compound your learnings. And then focus on learning velocity, not just running tests. The faster you iterate, the faster you learn, the faster you can make decisions, the better. And use that kind of HCE framework that we kind of talked about, which is the hypothesis, control, and evaluate.

And I'll add one thing. Bad tests are worse than no tests. So test with discipline.

Dude, that's a wrap. Um, that's episode 63.

Yeah, dude, that was awesome. Creative testing that actually works. How to, how to A/B test your emails. I don't know what we're going to call this title exactly, but We talked all things testing here. If you enjoyed this episode, please, please, please like, share, give it to your team members, give it to your favorite email marketer or marketer or even advertiser, because I think we dropped a lot of great golden nuggets on this episode. Thanks, Chase, as always. Make sure you check out all the cool links that we have below, including the vault, including joining our newsletter, including inbooks.ai. That's all my plugs here today. With that being said, thanks for being here today and we'll see you on the next one.

From the open to the click, they're the masters of the flow. Just watch your list grow and your leads turn to go. Did we retain you so far, or do we have to extend it? If you want results, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it.