Agility can often get framed as driving massive transformation—but sometimes it’s the milliseconds that matter. When every digital moment counts, small gains in speed and efficiency can have a disproportionately large impact on customer lifetime value and brand loyalty.
If you could speed up every digital interaction your customers have with your brand by a full second, what would that be worth to your business?
Today we’re going to talk about how even seemingly minor improvements in speed and performance can have outsized impact on customer experience—and revenue.
To help me discuss this topic, I’d like to welcome Jaxon Repp, Field CTO at Harper.
About Jaxon Repp
Jaxon Repp, Field CTO at Harper, has over 25 years of experience architecting, designing, and developing enterprise software. He is the founder of three technology startups and has consulted with multiple Fortune 500 companies on IoT and Digital Transformation initiatives. A partially-reformed developer, he understands what it’s like to wrestle with technology instead of benefiting from it, and believes passionately that if the Jetsons never had an episode where a config file error brought down the food-o-matic, it surely should not be a problem now.
Resources
Harper: https://www.harpersystems.dev/
The Agile Brand podcast is brought to you by TEKsystems. Learn more here: https://www.teksystems.com/versionnextnow
Catch the future of e-commerce at eTail Boston, August 11-14, 2025. Register now: https://bit.ly/etailboston and use code PARTNER20 for 20% off for retailers and brands
Don’t Miss MAICON 2025, October 14-16 in Cleveland – the event bringing together the brights minds and leading voices in AI. Use Code AGILE150 for $150 off registration. Go here to register: https://bit.ly/agile150“
Connect with Greg on LinkedIn: https://www.linkedin.com/in/gregkihlstrom
Don’t miss a thing: get the latest episodes, sign up for our newsletter and more: https://www.theagilebrand.show
Check out The Agile Brand Guide website with articles, insights, and Martechipedia, the wiki for marketing technology: https://www.agilebrandguide.com
The Agile Brand is produced by Missing Link—a Latina-owned strategy-driven, creatively fueled production co-op. From ideation to creation, they craft human connections through intelligent, engaging and informative content. https://www.missinglink.company
Transcript
Greg Kihlstrom (00:00)
Agility can often get framed as driving massive transformation, but sometimes it’s the milliseconds that matter. When every digital moment counts, small gains in speed and efficiency can have a disproportionately large impact on customer lifetime value and brand loyalty. If you could speed up every digital interaction your customers have with your brand by a full second, what would that be worth to your business?
Today we’re going to talk about how even seemingly minor improvements in speed and performance can have outsized impact on customer experience and revenue. To help me discuss this topic, I’d like to welcome Jackson Rep, field CTO at Harper. Jackson, welcome to the show.
Jaxon Repp (00:37)
Thank you very much for having me today.
Greg Kihlstrom (00:38)
Yeah, yeah. And I should say welcome back to the show because you’re a returning champion here. So good to have you here. For those that didn’t catch your last appearance here, can you give a little bit about your background and what you’re doing at Harper to help brands improve their digital performance?
Jaxon Repp (00:54)
Sure. I am the field CTO at Harper, which basically means I’m a grownup sales engineer that rather than not caring at all about the money and the revenue and the business model behind like solving the awesome problems we get to solve every day. have like least some awareness of the fact that we do need to be profitable and deliver return on investment for our customers.
Greg Kihlstrom (01:13)
Nice, nice. Well, yeah, let’s let’s dive in here and we’re going to talk about a few things today. But I want to start with what I touched on at the intro and just the speed, efficiency and experience that customers are really expecting and in some cases demanding. So in a world where 70 percent of consumers say speed impacts buying behavior, how do you think brands are and why are brands still missing the mark when it comes to performance?
Jaxon Repp (01:39)
I think like all things in business, there are points of diminishing returns where I can absolutely set up one server for every customer and personalize everything and make it super fast because it’s just literally you alone asking a database a question and getting back information and personalized content. That does not scale. It’s not any way to run a real business.
As you begin to do different things to appeal to your wide customer base, vertically scaling and adding resources to solve those problems in a non-distributed application format is always going to fail at a certain level. So the more we can distribute that load, the more we can do. And I think brands are rightly perhaps intimidated by the prospect of distributing that load and figuring out how to give an acceptable experience with an acceptable cost structure for that widely distributed audience base. So it’s not that brands don’t know what they want to do, what they want those numbers to be, how they want to personalize, what A-B tests they want to run. It’s how do I deliver that for an audience of
you know, a hundred million users, you know, spread around the globe and, and delivering, you know, a proper return on that investment.
Greg Kihlstrom (03:03)
Yeah, yeah. And so, you know, when we talk about agility, I mean, I work with a lot of large organizations on agility and, you know, transformation initiatives, these big picture, you know, large scale things. I think that’s what a lot of people have in their minds. And certainly there’s a lot of those going on and and and in progress. But what we’re talking about here is it may be large scale, but it’s made up of of
a few milliseconds at a time, right? So, you know, how can how can a few milliseconds change customer behavior at scale?
Jaxon Repp (03:38)
Well, in short, they add up, right? So every, every function from, from querying my shopping cart to delivering me a bit of content in the product description that is perhaps tuned just for me based on an advertising campaign or a cohort or something, you know, about me, all of those things add up to, you know, am I able to click the button once you finally convinced me and am I even still on the page?
because it took too long to load because it’s hard to make it perfect for every single person. So you begin to identify cohorts of individuals. These 10,000 of my 10 million users respond in this way. And how do you isolate, identify, deliver that content to them? And then when you are doing it 10 milliseconds faster or a hundred milliseconds faster, not even a second, you begin to look at
You know what those response rates are and Amazon and like lots of, lots of people have done very real studies that say for every X milliseconds, your conversions will increase by this much. What a conversion is for you. Obviously we’ll map to revenue. For some of our more sophisticated customers, they know exactly when we save, you know, a hundred milliseconds on delivering page content or, or, you know, customized content or AB testing. They know exactly how many.
millions of dollars that might make them over the course of a year. Others, it’s very much a lagging indicator. So is a more performance site that is indexed more rapidly by a search engine, I’m going to end up with a better score, which perhaps ranks ahead of a competitor. And now all of a sudden you’re third instead of fifth. And all of a sudden I’ve got a revenue uptick. And so it’s really data driven.
but you map it back ultimately to those performance metrics.
Greg Kihlstrom (05:33)
Yeah, yeah. And so when we talk about speed, well, let’s let’s talk a little bit about composability and, you know, modern web infrastructure. You know, we talk quite a bit on the show about about composability and, you know, just the benefits of flexibility and, you know, best of breed and all those kinds of things. There’s less conversations about speed when talking about composability.
How do you see performance fitting into the broader conversation about composability?
Jaxon Repp (06:02)
I think being able to assemble a host of functions that perform whatever the outcome is for your user is, it’s very appealing. You think back to every language ever has had the concept of, know, NPM has packages and Ruby had gems and like everybody has ways to things, make things easier.
Greg Kihlstrom (06:20)
Right.
Jaxon Repp (06:23)
But you get into inevitably like, this is the package I want to use, but it doesn’t do exactly what I want to do. So how do I build on top of that to make it do exactly what I do? And now you’re seven layers away from the actual core functionality. And I, don’t need everybody to write all their backend in C, but at the end of the day, what we focused on was we know we want to make something extensible. How do we allow that abstraction to not take away from the underlying
So when we built our platform, we combined the database, the API layer, the in-memory cache, the modern interfaces, HTTP or streaming. And we basically put them all into one package so that no matter how you composed those individual attributes of your project, they were all sitting directly on top of that persistence layer, which at the end of the day, the data drives everything.
We wanted to make sure that those individual functions that you write are as close to the bare metal as possible in a modern application structure where nobody’s really sitting on top of bare metal anymore. But we wanted to at least give you a head start toward a performance solution because lots of us can build prototypes that function, but as soon as you go to scale, everything falls apart. And that was really the problem we were trying to solve for.
Greg Kihlstrom (07:45)
Yeah, I mean, it seems like it’s a similar but different challenge. I mean, in a composable environment, you’ve got a number of different sources with a number of different perhaps approaches to speed and performance even, or just maybe different tolerances for that. So, I mean, it becomes really important to choose the right part as opposed to the little monolithic, you you kind of get what you get, I guess. But, you know, in a composable environment, you know,
Jaxon Repp (08:09)
Yeah.
Greg Kihlstrom (08:12)
speed and performance. mean, that should be a very important aspect of choice, right? In addition to features and things like that, correct?
Jaxon Repp (08:20)
Exactly. At the end of the day, we want to allow you to be as flexible as possible. We want you to just be able to write the function that seems to make the most sense. And when that function is directly on top of the data, there’s, might, we might argue about your for loop. we might argue about the structure of your reduce function. but at the end of the day, all of that’s going to be relatively fast running in Ram, but
How can we make sure that it plays well with all the other components, that they all interact well together, and that they all sit at effectively the same level so that I don’t have a hierarchical tree of functions that kind of cascade inefficiencies across my cluster?
Greg Kihlstrom (09:04)
Yeah. So there’s a lot of marketers and CX folks listening to the show. They’re very focused on customers. so, you know, they know customers want fast access and availability, but they’re also often looking at the features of an application, not necessarily at the speed and performance. You know, they’re assuming someone’s looking at that and that it’s it’s performance and all that. What should marketing and
CX leaders be asking their tech teams who are inevitably being charged with some of this evaluating performance, what should they be asking when it comes to ensuring both scalability as well as performance in a composable stack?
Jaxon Repp (09:42)
I think every developer at one point or another has built the perfect application that would have scaled flawlessly across the entire universe. And then the marketing team comes in and adds like 27 measurement scripts to your application. And you don’t feel like you’re getting everything that you built communicated. And all of these other things kind of sit out to the side. So I really kind of want to say…
Greg Kihlstrom (09:57)
Right.
Jaxon Repp (10:10)
And we ask this of our customers all the time. What are the real numbers that you are after? Because there’s a chance that a really well structured application can capture all of the metrics that you’re after. We’re not going to be a Google analytics, you know, you’re going to push stuff off. Everybody has a dashboard that’s perfectly formatted for what they want. But if we can be that kind of first principle source of truth, then we can push that raw data in and.
A lot of times there are need for a ton of client side stuff. If we can extract performance metrics, query metrics, user metrics from the backend and reduce kind of the pantheon of tools that often get applied on the client side. Because a lot of those rely on that browser to execute things in order and inevitably some of them some of the time that gets in the way of what that actual user wants, which is to simply click the button by the thing.
Greg Kihlstrom (11:09)
Yeah, yeah, yeah. The marketers want to see every little step in the process, but the customer just wants what they what they want. Right. It’s a and both are both are valuable. But but, know, to your point, when one shouldn’t get in the way of the other. And that kind of brings me to the I want to get back to the the the business case here and the business value of speed and, you know, a faster site doesn’t just feel better. Certainly, you know, we as as consumers know sites that load fast or others that we will do anything to avoid probably, faster sites drive conversion, satisfaction, loyalty. How are your customers translating performance gains into measurable business outcomes?
Jaxon Repp (11:52)
I think it comes in two phases. The first one is my lighthouse score is terrible. The CEO’s granddaughter tried to access the site this weekend and called her grandfather and said, why is your company terrible at what they do? And that’s a real example, by the way. And so when we are pursuing kind of performance, we are constantly pursuing, you
the right customer outcome, whether that is a revenue uptick, whether that is performance as granular as LCP so I can affect my lighthouse score so that I could possibly rank higher for a particular product when I’m competing against the really the largest retailers. Like I’m competing against Amazon. They have unlimited resources. They have servers everywhere in the world.
I know that they have a hundreds on all of their lighthouse scores. How do I compete against that? And it really is like, let’s do phase one, which is let’s get your core web vitals to be improved by X number of milliseconds. Let’s get you green at the very minimum. Then we start to look at what are all the other things you want to do to improve conversion. And a lot of times that has to do with A-B testing, serving multiple templates.
being able to serve different content for different users. All of that sort of personalization, which is incredibly resource intensive and hard to accomplish in perhaps traditional architecture, where our multi-discipline platform, our composable application platform allows you to accomplish that with horizontal scale so you can remain performant.
do all of those A-B tests, do all of that personalization and still do it in a way that’s cost effective. Because again, I could assign every one of my customers a server and that would be perfect. I could do anything I wanted and everybody would be as fast as they were the very first time you visited your very first website because you were the only visitor. So that’s what we’re trying to do is figure out what’s that first thing we can do and then what are those bigger restructuring.
Greg Kihlstrom (13:46)
Right.
Jaxon Repp (14:01)
projects we can do to help more, you know, more performance, to squeeze more performance. Let’s get that out of there. I hated that answer.
Greg Kihlstrom (14:10)
sure.
Jaxon Repp (14:11)
Could you hit me with that question? I have a much more succinct answer now. I realize I got through that. Yeah.
Greg Kihlstrom (14:14)
sure, sure. Yeah, of course.
So a faster site doesn’t just feel better. It drives conversion, satisfaction and loyalty. How are your customers translating performance gains into measurable business outcomes?
Jaxon Repp (14:25)
I think in many ways we look at that as a two tier process. The first is kind of what do SEO bots see? Because for a lot of them, it is where they are ranked, which is their number one concern when they come to us. It’s my page is slow, but it’s negatively affecting how consumers find me. Not even to the point of how they experience you. So what can we do to improve core web vitals like LCP?
time to first bite, those things that will get you ranked higher. And those are very easy for us to do as just kind of that front layer. Next are the things that truly impact conversion, like personalization, A-B testing, and that is a, it’s a heavier lift, but our platform is designed to allow you to accommodate all of that personalization and testing for groups very efficiently by providing everything you need in one box.
And, and deploying that, even though it is, you know, a little bit of porting your application potentially, or moving your APIs onto a distributed infrastructure. We try to make that as easy as possible because the real solution for scale is horizontal and the real solution for personalization and the real outcomes of feeling like that site is truly targeting me and giving me everything I need is some sort of infrastructure that can do that on a cost effective manner.
at scale for users that might be distributed across the entire world.
Greg Kihlstrom (15:51)
Yeah, yeah. And so maybe to ask the inverse of that question, but you touched a little bit on some of this already. But as is probably always the case, teams are being asked to do more with less. And so they’re being asked to prioritize. And a lot of times, the things that get prioritized are like the new fancy features and the new not necessarily
making a site more performant and, all those kinds of things. What’s the cost of inaction here, you know, of simply saying, well, yeah, we’ll, get to that. because we want to, we really want to launch this other thing or whatever the case may be. What are brands risking if they don’t make performance a priority?
Jaxon Repp (16:31)
I think the first is that you are judged relative to your competitors and everybody is trying to make things faster. They are also trying to launch new features and new personalizations and including AI generated content and all of the other things that they’re trying to do. And historically what we’ve seen is the old school sites that are really, really plain. If you look at Amazon, when you are searching,
for any given thing, their results are very good because those pages are very simple. They are very easy for a bot to index and they are very performant and easy for me to use. When you are attempting to compete against an unlimited infrastructure kind of competitor, you are tasked with a lot of things. You have to cash, you have to distribute, you have to ensure that you have
you know, multiple layers of performance enhancing attributes as part of your solution. and everybody is doing that at the same time. So I would say as you are trying to add all of these new features, inevitably what occurs is it feels really good for a while. And then maybe you get a little traction. Maybe you become slightly more popular. Maybe there’s a new feature that everybody loves and they all go.
If that feature doesn’t scale, if that feature slows you down, you’re behind actually where you were when you started. And that is usually when people end up coming to us, they say, this was the most amazing idea ever. Ultimately, the execution fell down because of the stack we used or the paradigm we kind of conceived of, or we were operating, you know, with infrastructure choices that we made five years ago, and they’re not built to support that.
Can you help us? And that’s ultimately where we play is helping deliver a more composable, intelligent performance stack to, again, allow you to scale those cool features to address that wider audience and to ultimately deliver that return.
Greg Kihlstrom (18:35)
Yeah, yeah, love it. Well, Jackson, thanks so much for all your ideas and insights here. One last question for you before we wrap up. like to ask everybody, what do do to stay agile in your role and how do you find a way to do it consistently?
Jaxon Repp (18:48)
I think being kind of the customer facing, you know, sales engineer, it is always an interesting challenge. All the challenges are different. They certainly play in the performance realm. They often play in the digital commerce and conversion realm, but every problem comes with its own unique, unique challenges. So you are by definition agile in this role. And I think as an organization,
We are constantly looking at the variety of challenges issued to us by our customers and trying to stay one level abstracted above that and say across all of these customers, how does our platform in its uniquely composable paradigm, how could it address many birds with one stone? What are the common threads that we see? What are the common features that we would develop?
that would allow us not just to solve the problem for this one customer, but perhaps for two or three or 10 or every single one of them in a way that makes sense for our product, makes sense for the customer’s ultimate outcome, but really and probably crucially for the developers that are going to be doing this implementation because we were all developers who built this platform to solve the problems that we were seeing.
And ultimately, it is a tool that developers will use to create those solutions for their customers or for their companies if they’re internal. And that is what we’re really focused on. The best developer experience and the most performant outcome for any application at any scale.
