On this episode of Performance Delivered, we welcome guest Keith Hagen to talk with us about powering incremental improvements with insights.

Keith is the Vice President of Digital Marketing at SelectHub. SelectHub offers a streamlined solution to company stakeholders for enterprise software product evaluation, vendor sourcing, and IT procurement.

Keith says, “getting the customer insights as to what our audience wanted helped us narrow down what we needed to do.” Listen as he shares:

  • A dirt-cheap, simple do-it-yourself way to garner customer feedback in 2021 – it’s a little bit old school and really pretty simple
  • How to get better (and faster) insights and make the right incremental improvements to your website
  • Three sure-fire ways to take your insights to the next level
  • And much more

Mentioned in this episode:


Voiceover: This is Performance Delivered. Insider secrets for digital marketing success with Steffen Horst and Dave Antil.

Steffen Horst: Welcome to the Performance Delivered, Insider Secrets for Digital Marketing Success podcast, where we talk with marketing and agency executives and learn how they build successful businesses and their personal brand. I’m your host, Steffen Horst. The topic for today’s episode is powering incremental improvements with insights. Here to speak with me is Keith Hagen who is the VP Digital Marketing at SelectHub a technology selection management solution that eliminates the messy error prone business of IT and software requirements compilation, vendor evaluation and sourcing by centralizing all activities on an easy to use intelligent platform. Keith is a growth marketer with 18 years of experience driving online growth through strategy, demand, engagement, acquisition, conversion, and customer loyalty. Keith, welcome to the show.

Keith Hagen: Hey, thank you. Thank you, Steffen. It’s great to be here.

Steffen: I’m happy to have you too. Well, listen, Keith, before we start to talk about today’s topic, I’d love to find out more about you. Tell our listeners on how you get started with your career in particular, how did you end up with digital marketing?

Keith: Sure. Okay. So, um, let’s see, I did my, you know, University at Carleton University in Ottawa, Ontario, Canada, and I, in psychology, sociology. Found that left me without, you know, any real skills to in the job force, of course. And in a typical Canadian fashion, I did my kind of a postgraduate program to learn some skills. And from there that led me into programming which quickly, you know, led me to a visa to come to the US. And I worked as a programmer full time for about five years. And at that point, the internet was so new, that when a job opening, I was working at a place called Gates Corporation, previously known as The Gates Rubber Company. 

And at that time, was the largest company headquartered in Colorado. And I was working there, a job opening came up for basically running digital marketing at the company. And because I was really the kind of the lead internet developer for the company at the time, they asked me to move over to the business side and run all things digital marketing. And so you know, there I was, at a fortune 500 sized company, they of course, you know, we’re headquartered outside of the US, in the UK at the time. But there I was running this, running all things internet related at at a fortune 500 company. And I was really just by myself. 

We had some outsourced agency resources, but you know, might have accumulated to like one and a half people helping me. And I did everything. And you know, a few years into that position, got recognized by B2B Magazine, one of the Korean publications is having the one one of the top 10 b2b websites, which was quite, great, because we’re being compared to all these companies that were 10, 20 people. And here we were less than three being recognized for, essentially, you know, bringing this this huge group of companies online at a, you know, that pivotal time back in the, you know, early 2000s. So.

Steffen: Yeah, well, interesting. Now, with a small workforce, you must have done something, I don’t know, different. Something special in order to achieve that recognition. What was it that that allowed you to kind of achieve results with a small team from which other companies needed many more people?

Keith: You know, I think it was that we built systems. And this was before there was a CMS. So we built our own. We also built our own email marketing automation tool. And I have to shout out 90 octane, who’s now a sizable agency in the Denver area. It was it was just three of them helping us at that point. And it was really just building systems that we could leverage and also getting the getting the customer insights. We were blessed with a large brand. And lots of people were coming to the website. But I think by really getting the insights as to what our audience wanted, really helped us narrow down what we needed to do. So I’ll give you an example. Drive design engineers, so you know, design engineers, they’re just trying to make stuff and they need, they need AutoCAD drawings. 

And so you know, getting the insight that that’s what they’re looking for, you know, we put the design drawings, the AutoCAD up on the website before anyone else did. What that did is it allowed it allowed all these all these design engineers to, you know, create what they were creating with our parts. Which essentially baked the Gates parts into all of these new applications. Like the Harley Davidson, you know, drive chain, that the rubber, the rubber belt system that they used. Or sail boats, or electric bikes, or, you know, bicycles. Whatever it was, all these new applications, found it easier to work with Gates than it did with any other company. And that changed the dynamic at a time when nobody was really expecting any leads to come in line, you know, coming from online. 

You know, within a few years, we’ve after we get, you know, finally got a proper CMS system in, we found out, you know, 96% of the leads were coming from on what were originating from online. And quite a pivotal time back then. But that’s it was really the insights kind of insights driven, you know, is what we’re benefiting from was really knowing our customers. And we would, you know, that’s back in the day where, if I saw somebody fill out a form, I would call them and say, you know, what are you looking for? What can I help you find on the website? Things like, navigation was brand new back then. You know, nobody knew whether a left nav was going to be better than a, you know, front nav. So we were figuring all this stuff out. And but it was really building systems and being insights driven.

Steffen: Yeah. Well, obviously today’s, today’s topic is powering incremental improvements with insights. And you just started to talk about insights. Now, how do you drive actually continuous incremental improvements?

Keith: You know, the first thing is remembering that that’s what you have to do. There’s been for a long time a redesign culture in the US, especially pertaining to websites. I would say, from the time that everybody started building websites, they were just rebuilding websites. And there’s some periods where redesign was necessary. When we all moved over to responsive websites, for instance. That was a good time to do a redesign. But outside of that, there, you know, it’s really been a redesign culture. And it’s the websites for the longest time were being driven more by design resources than they were business resources. And it took a while for the business acumen to catch up to online. And for there to be people like myself who were results, business folks focused online, and could see through the smoke and mirrors of design or SEO or social or whatever new fad cropped up. 

Some, you know, somebody who could understand how to, you know, what it meant how to leverage it, what did or did not need to be done. And that’s the long, you know, that’s kind of the long and short of it. Just from the point where I started working on the websites that I owned. And with Gates Corporation. In the early days, I inherited about seven sites. And it was impossible with the resources to redesign anything. And we really didn’t have the budget, the business didn’t believe in the internet. The executives at the time thought that it was just unnecessary to have a presence, they didn’t really see its value at first. You know, until that value is proven. So there is no money for a redesign. And everything had to be incremental. And that’s really where I you know, I started the the methodology and, and luckily being at a manufacturing. 

And, you know, I, I have to hand it to the Gates Corporation, they train their people. They, they educate them, and I was you know, the beneficiary of courses in Six Sigma and, and lean manufacturing. And I got to take those principles and apply them to what I was doing online to the point where I was even participating in voluntary Six Sigma projects of for online even though I was really the only one who understood the internet at the time at the company. And I didn’t really have any checks or balances are mentors. But I was able to follow through these these Lean Sigma methodologies and improve the site. And so I think that that was really kind of the core for me.

Steffen: I couldn’t probably agree more with you as it as it relates to website improvements being more driven by design. I remember, I used to have a client back in London when when I lived in London, it’s a it’s a big luxury company. And, you know, they they wanted to stand out and they built a website that was heavily video focused. Now, as you can imagine, that created a lot of problems at the end of the day. Content wasn’t readable site was slow. It was confusing for for for customers that wanted to buy their products. It looked beautiful. But from a usability perspective, and all the other elements that actually make a business impact. 

It was just terrible, you know. So they had to do another, another run at redoing the new site in order to make it work for what they wanted the website to do. Now you obviously we’re talking about website redesigns. Where do you take information from to make decisions on what elements of a website to improve? Because that’s what we’re talking about. We’re talking about, you know, making incremental improvements, not kind of throwing everything overboard, and starting from scratch again. So where do you get that information from? And what type of information are you looking for?

Keith: Yeah, and the place you get the information from is really your customers, and you have to get as close to them as you can. And how you do it doesn’t matter. I thought that I think that was the biggest epiphany for me, early on, where I was, you know, in charge of this large initiative without many resources. And I had to get questions answered. Now, I did have some online, I guess you could call them mentors. I didn’t get to spend much time with them. But you know, Amy Africa, for instance, was a big influence of mine back then, I realized that she was on the speaker circuit talking about conversion optimization before it even had a name. And like I was doing with all types of marketing back then, you know, she was basically educating for free people on conversion optimization, email, advertising online. 

All the things that I was doing with Gates, she was out there talking about, and I got to, you know, through through people like her and UX experts, all the all the notable names from from back in the early 2000s. You know, through them, I got to have read, read about people having similar experiences. And the big takeaway was, you can just call people on the phone. Ask them, how is your website experience? And in the early days, we were asking, were you able to find what you were looking for? Now, today, hopefully, that’s not as much of a question. I can tell you, it’s for sure, still a question. 

But hopefully, it’s not as much the focus anymore, because back then people were just, you know, really new to the internet. But you can, you can still call your customers, you can still pop up on the chat, I remember when liveperson first came out. And when I was looking at the service people at one of the companies I consulted for, and I saw that they were just popping up and asking people questions, and I’m like, well, ask him if they can find stuff. Ask him if, ask him if you know, and I put up this battery of questions, and gave him a spreadsheet, and I had all these customer service people, peppering these questions and getting these answers that gave us a beautiful matrix as to the user experience for that particular site is about finding auto parts. 

So you can you know, using chat, one of my go to favorites has always been user testing. And I was one of the first users of usertesting.com. And that was right from the beginning a phenomenal resource. And it still is. The user testing if you create a good scenario, and you find the right audience, often recruiting you know, your own site users. And again, you can use chat for this, try to get them into user testing. You get amazing insights just by watching people try to achieve something that you think should be elementary. And it’s humbling, it’s probably this line of work is probably the most humbling thing you could possibly do. 

When you put so much effort into launching a site or making a change or putting together a homepage, and realizing that it just doesn’t work well enough. But you’re never too far. You know, it’s, you know, we can talk about A/B testing, which is a great way to get insights. But more importantly, it’s a great way to test the ideas you get from those insights. And if you get a good insight, it’s going to compel you to test and you really shouldn’t test anything that you’re not compelled to.

Steffen: How important information from systems like hotjar, crazyegg, or the data points that you collect for your web analytics system. How do you how do you use those information?

Keith: Yeah, I, I think hotjar is great. I think it’s still has a ways to go. I love clicktale. And if you put the time and energy into configuring it, right, it’s, it’s a, it’s a beautiful thing. But whether it be hotjar or clicktale, or some other screen recording system, screen recordings are only good as the context that you’re watching the recording in. Is it a new visitor? Is it a returning visitor? What do they look at before? What did they look at now? You need to understand the full picture. You can’t just watch a screen recording, recording in isolation, and know anything, it’s useless, it’s garbage. You really need to have the context of what does that user trying to achieve. 

And so for example, if you’re going to do screen recordings, you really need to design your site in such a way that you can identify people simply by navigating. If you can call out an audience from a navigation point, let’s say your homepage calls out a particular audience. Let’s say it’s, you’re an auto parts seller. And you call up mechanics. You know, if you say, you know, mechanics start here for your part search. From that point on, you’ve got the context, you know that it’s a mechanic, you know, they’re looking for parts. That’s where screen recording becomes useful. Because now you’ve you’ve identified people right from the landing page, that they that they arrived at, all the way through, through cart and checkout. 

That’s where screen recordings are important. But if you don’t design your site, in such a way where you can easily, you know, identify people, screen recordings, definitely is going to have its challenges. I’ll give you an example of screen recording success. Back when I had my consultancy, Conversion IQ, I took a vacation. And on that vacation, I finally had the time to go through screen recordings of my of my largest client, which was a national pest control company. And, you know, I’ve watched probably 30 hours of screen recordings that week. And I was able to do it in isolation, I took good notes. And I got to identify patterns. And this is something I’d done before. But it’s so time consuming. And this is really important to talk about. It’s your insights, you want them to be good. 

You want them to be fast, and you want them to be good, cheap. screen recordings are good, and they’re cheap, but they’re not fast. So, you know, after 30 hours of watching hundreds of screen recordings, I realized that people who were looking at bedbug infestations, they always follow the same pattern. First, they wanted to identify what it was. They wanted to identify it as a bedbug. Second thing they wanted to figure out was, how did they get it? The third thing they wanted to do was figure out how serious it was. And only after that, would they actually try to remedy it with any of the calls to action on the site. Now, if you took took a look at any pest control company, any website, they always help you identify what it is. 

And then they offered a solution. At no point did and did any site, help people figure out how they got it, and how serious it was. And once once those two aspects were integrated into the site, the conversion rate went went through the roof. Number of people working in the call center with were tripled over the years simply because of this, the shift and understanding that came from screen recordings. So screen recordings are definitely powerful. But you’ll get the insights, they’ll be good, and it will be cheap, but they won’t be fast. Unless unless you design your site upfront better. So

Steffen: Yeah. Now then, you know, adding on to what you just said, what are the best insights, you know, given, you know, you have limited time, limited resources, and money? Can you call out a few or that can people can do companies can do that. That can provide a lot of insights, but don’t cost that much don’t take that much time etc.

Keith: Yeah, absolutely. The, you know, first one is user testing. You know, there’s a lot of lot of vendors out there, I’m not partial to any one of them. You know, try my UI or usertesting.com are are good resources. That’s going to be the place to start. I think everybody should do that just to orient themselves. Get five user tests and give people their you know, your primary use case and just watch people use your site, that alone is going to spark, maybe even a redesign. But it’s probably going to spark a lot. After that, you know, I’m a big fan of using chat and, and recruiting people to, to talk to. And when you actually talk to your customers, it gives it a different, a different realm. And it’s, it’s gutsy, you have to not be, you know, willing to give something away. 

You know, something, say, hey, for 50% off your purchase today, would you be willing to walk through the site with somebody. And, you know, as soon as they say, yes, you know, zoom meeting goes out to them, and you guys get on a zoom call, they share the screen. And, you know, you’re shopping with somebody. Those are the, in my experience, the two best ways. When it comes to insights, you can also use surveys, pop up surveys, polls, they need to be really targeted, I find them far, far better than any kind of long form survey. But if you you know, hotjar, of course has those polls. Qualaroo will give you a little bit more dexterity, you might say with those polls, but your polls are going to, you know, give you a lot of context. 

And I love using audience polls or buying cycle polls, just just to help me understand at what stage the people using a page are for. Let’s say it’s a key navigation page that helps you sort people through various categories or products. Knowing what stage the people who use that page are on is priceless. You know, are they in the research phase? Are they in the consideration phase? Or do they expect to purchase within 24 hours or even purchase you know, within the hour? Those are the type of questions you can poll people on and and get that context that you’re missing about that page. Because until you know what phase people are at, you know, you won’t understand. Likewise, in terms of mindset, you can use those polls to figure out, you know, what the user mindset is? 

Are they again, are they going to make a quick decision? Are they more focused on the best solution, versus being more thorough and analytical and you know, where they’re going to do copious amounts of research. And so, when it comes to mindset, and buying cycle and audience type, you know, just who the audience is, if you’ve got more than one on your site, that’s where pop up polls are awesome. And that will add a lot of insight into your heatmaps into your screen recordings into your Google Analytics, or other analytics that you’re using. And will also make for a better user tests. Understanding the mindset, the buying cycle and the audience on any given page, any starting point, that’s where your insights are going to start.

Steffen: Well, now that you have collected a lot of insights, how do you prioritize them? How do you prioritize the insights in regards to what to take action on first?

Keith: Sure, yeah, I you know, the action, you should probably take from your insights is testing. A/B testing. I’ve personally done over, tested ran, I’ve run over 4000 A/B tests now. And it’s not bragging. It’s it’s a dubious honor of running so many tests, but that should be your action is running a test. And you should be from your insight, you should prioritize, you know, in three basic ways. One, what’s the potential? Two what is the viability of the test? Can it be tested? And can it be implemented by engineering or your developer group? And then third, is it aligned with with executive goals? So those are the three things you want to look at. There’s a great free tool out there, called illuminate from Brooks Bell agency. That’s another CRO agency that a, competitor to to the one that I used to have. 

And illuminate is a tool that allows you to gather insights, prioritize them, basically on those three things, I think, and then, and then kind of prioritize and track your tests. And that’s what you’re looking for. You’re looking for, you know, what’s going to have the biggest impact, what can and what can actually be done, what’s the difficulty of it. And, you know, I would score it in in such a way where you give the impact score of one to five. You give the the ability to be done a one to five, and you give the importance to the executive group a one to five. And then you take the impact, give it a multiplier of say, five, the abilities to do the work of that a multiplier three and then whether or not it’s it’s core to the objectives of the executives, give that a multiplier of two. And you’ll come out with a score and with that you’ll be able to score every initiative that you have.

Steffen: Now that you kind of identified which insights to take action on, and then how to prioritize them, once you launched them, how do you measure the improvements? How do you measure which improvements have made the biggest impact and which improvement you potentially want to continue to pursue to to even find out greater value?

Keith: Yeah, great question. Like I said, you really want to your insights to be tested. You know, I look at things that I’ve tested. And when you get an insight, it’s probably going to go well. Your test is probably going to win. If you’re driving tests, A/B tests, multivariate tests on insights, your win rates going to be 80% 85, 90% even. The problem is you don’t know what that 20, 10 to 20% of losers are going to look like. They don’t know what the impact is going to be. It’s very easy to think that you’ve got a great winner. And you maybe well do. Recently, I had an excellent winner on a test, it did exactly what I thought it would do. However, it also cannibalized other goals. And those goals were something that were just more profitable, and had a higher success rate, you know, through the full funnel, with the SDR. So just because you have a winner doesn’t mean that it doesn’t come at a cost. 

And you want to test everything. You want to understand that every dynamic and testing isn’t something that is you wouldn’t rush. I think the biggest problem with testing is that people don’t run a test long enough. They believe the numbers without putting a proper methodology around the testing. I’ll give you an instance. When you start to test who did that effect? Because anytime that you change the website on a returning visitor, they’re going to notice and there’s going to be an impact. Let’s say you have a homepage and you had a call to action, you try a new location. But for some people come back, if that call to action suddenly moved, they’re going to be confused, it’s going to add friction, it’s going to cost you sales. 

So when you start a test, you have to be very careful to make sure that you only start the test for new visitors. Returning visitors shouldn’t see it or else you’re going to pollute the results of your test. You also have to know the type of site you are. If you’re starting a test on Sunday, and you know, your audience starts shopping on Saturday, you’re probably going to miss you know, and then you run a test through Friday, you’re going to miss the you know, your main your main people who basically won’t have gone through your test. So now you’re, you’re not actually testing them. You know, the, let’s say you sell children’s bicycles, you know, most of that traffic is going to hit on Saturday, Sunday during the weekend. 

If you start your test on Sunday night, you’re going to miss most of that audience. And you’re only going to get the kind of parents who are shopping for their kids during the weekday, not during the weekend. And so that’s, that’s also important. And when you when you stop a test, you also if you stop it cold turkey, you’re not going to know how people in the test finished. And often buying cycles go on for weeks. And if you don’t let a test run out, and so you can gather all those people, you’re not going to see the results, the returning visitor results in your test. And that I’ve seen that totally change the test results. 

So the problem with testing is that people don’t follow the proper methodology. And I feel really passionate about this. I wrote a book about it. It’s up on Amazon if you if you search my name on Amazon, and I don’t profit from that book at all. It’s it’s sold with my company, when I sold my company, but but explains the whole methodology. And I think the books like 99 cents, I hope it’s of use to people. But you really, testing is something where when people test and then they implement, they don’t see the changes that they test. The test might promise them a 20% conversion rate increase, and then they implement and they don’t see any increase at all. Well, that’s simply because the test wasn’t run properly and it wasn’t reported out on properly. And so you want to test everything significant on your site, but you want to test properly.

Steffen: Keith, thank you so much for sharing some great information here with me and the audience. If people want to find out more about you and SelectHub, how can they get in touch?

Keith: So LinkedIn is probably the best way to reach out to me. My email is Keith.hagen@gmail.com.

Steffen: Thanks everyone for listening. If you’d like the Performance Delivered podcast, please subscribe to us and leave us a review on iTunes or your favorite podcast application. If you want to find out more about Symphonic Digital, you can visit us at Symphonicdigital.com or follow us on Twitter at Symphonic HQ. Thanks again. See you next time.

Voiceover: Performance Delivered is sponsored by Symphonic Digital. Discover audience focused and data driven digital marketing solutions for small and medium businesses at symphonicdigital.com