Does your sales process line up with your buyer’s journey?
In this episode, you’ll learn why addressing your customer’s concerns during the buying experience can reduce friction and boost conversions.
Tune in as I chat with Barron Caster, the Director of Growth at Rev. He reveals the methods his team used to triple Rev’s conversion rates over the last three years—and how you can become a better marketer by applying these principles to your business.
listen to this episode
- Why inexperienced marketers focus on selling the product features
- Where to get started with identifying your customers’ concerns
- Barron’s favorite place to find out how to improve conversion rates
- What happens when you only focus on the leads who don’t convert
- The surprising reason why you can’t rely on accurate tracking data
- Exactly what type of questions you should ask customers over the phone
- How to sort through your research data (and create a hypothesis)
- The pros and cons of running A/B tests individually vs. group testing
Louis: Bonjour, bonjour! And welcome to another episode of EveryoneHatesMarketers.com, the no-fluff, actionable marketing podcast for marketers, founders, and tech people who are just sick of shady, aggressive marketing. I’m your host, Louis Grenier.
In today’s episode, you will learn how to reduce friction within your buying experience by addressing customer concerns and customer objections. My guest today is the Director of Growth at Rev.com, which is an amazing transcription service I use.I’ve been using with Everyone Hates Marketers at least for the last year and a half. When I say it’s amazing, I genuinely mean it. And I’m not endorsed in any way, shape, or form by Rev.
My guest has been with Rev.com for three years now at the time we are recording, we’re going to publish this episode. He moved from Associate Product Manager to Product Manager to Senior Product Manager in Charge of Growth to Director of Growth—managing the entire growth and marketing team at Rev. So he knows a few things about product and marketing.
In the past two years, he and his team have more than tripled the conversion rate for three different products that Rev owns. They learned nice things that I talk about a lot on the podcast, which is they moved from the shitty growth hacks to a clear process to truly improve the buying experience of users.
Barron Caster, welcome aboard.
Barron: Thank you, pleasure to be here.
Louis: Part of my experience in the past in my career, I used to do a lot of conversion rate optimization. I even had my own agency. One of the things that keep coming back as, not necessarily a quick-win, but something that works really, really well is to genuinely just answer people’s objections or concerns on the page.
Knowing what question they have, answering that before they even start thinking of it and just going for it. That usually increased conversion quite a lot, because it’s as if you’re reading their mind.
I’ve answered a bit of the question I wanted to ask you first. From your perspective, why is it important to answer customer concerns throughout the buying experience?
Barron: Yeah, it’s incredibly important, because you’re trying to reach someone and let them know that you have a solution to their problem.
What inexperienced marketers do is tell them why your product is amazing. They just want to talk about all of the features they have, or how incredible they think their product is—and no one cares. You have to appeal to people where they are and solve their problem with your product.
By figuring out what questions people have, what they’re looking for, and by proactively answering that you relieve them of that stress. It’s not someone hawking something to me. It’s me finding something that will make my life better.
Louis: Yeah, and stress is an important word here. It’s searching for a solution, like searching for a solution to your problem could be quite painful or stressful for people. That makes sense as well, answering the objection this way, with this state of mind.
Why do you think marketers tend not to do this? Why do you think they keep talking about how amazing their product is instead?
Barron: Because it’s a lot easier. A lot of people know their product. They know the company they work for. They know all the features they have, and they’re embedded in an organization that talks a lot about their service.
It’s a lot harder to get out of the building and learn from your customers. To talk to your customers, figure out their concerns, what they’re looking for, how your product or service fits into their life in a seamless way, and how you solve their problems. It requires a lot of work, and it’s not intuitive up front. At first, you think, “My widget is better. Everyone should use it.” Then, you evolve to figuring out, why should people care about this at all?
Louis: Let’s play a little game together, and I think this game is going to last 40 minutes or so, which is imagining that we work for a company. We are consulting for a company that have conversion issues or growth issues.
You can use obviously the experience you have with Rev, and the way you’ve done it. But they struggle with that. They struggle with growth. They struggle with conversions, and they probably do what you used to do at Rev.
Which is playing with growth hacks a bit that were not necessarily linked to each other or whatever. Let’s teach them to identify concerns and answer them in a coherent manner. What will be step number one? Where do you actually get started from in this process?
Barron: Great question, and I’m going to give you an abbreviated answer now. I’ve done a little bit of writing about it on Medium to go into details. But I always start with learning. You have to learn everything about your customers, your traffic, if you’re worried about conversion. How are people getting to your site? Why are they coming to your site?
I would implement a number of learning tools. In addition to just reaching out to customers you already have, I would add things like tracking to the page that you care about, your product or service page.
Like Mixpanel that tracks very granular clicks, so you can see what is actually happening as people come to your site. There are tools like FullStory to see how people interact with your site, what they stop and look at.
There are tools like HotJar to see where are people scrolling over, and what information is most relevant to them. There are tools like Intercom and Drift, where you can talk with people in real time while they’re on your website and answer questions.
My favorite place to go to find out how to improve conversion is the people who have already converted into customers. Figuring out why are they customers, what appealed to them about your service, what do they love most about your service?
And then, taking those words from your existing customers and using them to explain your service to new customers. You’re nodding, so I hope you agree with me.
Louis: Yes, if I was disagreeing, it would be a very, very weird way to do that. Yes, especially, to your last point, and we’re going to drill deeper into what you just said and try to deconstruct it.
Louis: But what you said at the end is critical. That’s something that I used to do when I was in conversion rate optimization—which was bad—which is people tend to focus on, why are all those people not converting?
They overly focus on those people who are not converting, obsessing over those ones, instead of obsessing over the ones who convert. Because, as you said, those are the ones who end up paying your salary and the bills.
They know why they converted, and they also know why or what almost stopped them from converting.
Barron: Yeah, and they’re more resilient. They really needed your service. If your conversion flow is horrible, and they still jump through all of those hurdles, they have a burning need.
Whereas for a conversion, you’re trying to optimize for the people on the margin. Figuring out what that core need is and relating it to the people who may not be as sure will really help improve.
Louis: Yeah, that’s a nice way to put it. Let’s deconstruct what you said. You said the first thing is you learn. You learn more about your customers, and because we are marketing we do market stuff. We learn about our market, which is the number one thing that marketers should be doing.
Let’s deconstruct those things together. You talked about a few tools. You talked about tracking. You talked about heatmap, session recordings, understanding what people do on your website. You talked about talking to customers.
Let’s bring that down to another level. What would be the steps required to learn more about that, about those customers?
Barron: In terms of the actual steps I would take?
Barron: I would, first, fully define what conversion means in this scenario. Wether it’s someone buying something off of my website or downloading a paper or installing a podcast, downloading a podcast.
And I would track every step possible that I know someone would take to get to that. Define your funnel as granular as possible. I mean, tracking clicks and form fills, every single step and finding out, where is the biggest drop off?
Why are people leaving? Where are people leaving? Then digging in with the qualitative and figuring out, why are they leaving from that step?
I usually start with the quantitative to assess where is the biggest area of opportunity and then hit it with qualitative tools. Whether it be the session viewing, where I’ve watched over 100 hours of people using our website on FullStory to HotJar.
Where it bubbles up a lot of those insights to a higher level. You can see where the heatmaps are, and where people are spending most of their time.
Louis: First thing, you define your funnel. As a service company, it’s traditionally homepage, product or pricing, sign-up, onboarding, and then up to this aha moment, the first value point.
For Rev, it could be when they upload their first audio file maybe, and so you map that out. In this podcast, I’m not trying to go too much into the actual tool, because I believe that tools come and go. And that’s fine.
Louis: But processes and the ideas behind them are super important. Yeah, Mixpanel, Google Analytics; you have plenty of tools that can do funnel mapping, but what you are saying is super critical.
Which is, how are you supposed to optimize anything or to grow anything without you having the full picture in front of you? That’s what you’re doing as step one. You would map that out, making sure that your tracking is in place, because a lot of times tracking is a bitch.
Barron: Laughs. We’ve all been there, yeah. I’ve made so many tracking mistakes. It’s incredible. I just want to say one key point. No tracking solution is perfect. Never expect that you’re going find the holy grail of tracking and everything will work out.
Know that your tracking solutions will be directionally correct, but the numbers are not perfect. I just want to throw that out there, because I’ve seen so many people make critical mistakes through trusting the tracking entirely.
Louis: No, no, give me an example. That’s interesting.
Barron: Mixpanel, for example, doesn’t work on some older browsers or browsers that have ad blockers. You think that you have a certain volume of traffic, and that they’re taking a certain action, but that’s not the case. It doesn’t capture the full story.
What we use as our source of truth is our revenue numbers, and then we work backwards from there. We use multiple tracking and analytics tools and bounce them off of each other. We use Google Analytics, and we use Mixpanel.
I know you don’t want to get too much into the weeds on those, but the idea is to have multiple sources that you can triangulate what the real answer is.
Louis: That’s an excellent point. Because I believe those technologies come and go, as I said, but the principles still are relevant. The key thing here that I completely agree with you on, is also the fact that digital marketers or people in tech and growth tend to overly obsess about numbers.
To the digits, like we know we have a 9.4 conversion rate. We need to increase to 9.6. While they forget a bit about the people responsible for those numbers, which are the actual customers behind their screen.
Most of the time what I’ve figured is, yes, you can look at numbers all day, every day. And get granular. But unless you talk to people, unless you really observe them, see what they’re actually doing and ask them—you’re never going to get to this level of insight that truly enables you to grow exponentially like you’ve done.
Not exponentially, because I think that would be a lie mathematically, but at least growing significantly.
Okay, so go back to your steps. You map out the full funnel, make sure that your tracking is in place, nothing is perfect when it comes to tracking. You’ll have at least a decent idea of where your biggest drop-offs is, and this is the second step for you?
Barron: Yup, yup.
Louis: How do you select the drop-off area, the area, the step in the funnel that is the one that you need to prioritize?
Barron: I’d love to say I have some magic formula for it, but the truth is that generally everyone’s biggest drop-off is at the first step of the funnel. It gets smaller as you go deeper on, because people have more intent as they move throughout the steps.
Every funnel I’ve ever seen has the biggest drop-off at the first step. I would start there. That’s usually where people are most unfamiliar with your product.
It’s their first impression of what you do, and where you have to speak to their questions and help solve their problem as quickly as possible. It’s the toughest to do, because you’ve built no trust with them. They really need to understand what you do in a quick amount of time, because people make decisions fast.
I always start at the top and then work my way down the funnel. Unless there’s something drastically broken somewhere, in which case the data points that out, and you jump there, fix that, and then move to the top.
Louis: Usually, in the main funnel for any company; it could be e-commerce, SaaS, or any other, it seems like the homepage is usually the hub, the start of the funnel, right?
Barron: Yeah, it’s either the homepage or a specific product or service page/
Barron: Depending on what they sell.
Louis: You said it very well. That’s because people in the first step, everyone can start in this first step. It could be a student researching for their thesis about all the competitors in the space. It could be a competitor looking at what you do, it could be just someone who landed on it by mistake, or it could be someone interested.
As you said, after the next few steps, once you go from the homepage to a product page, you start to show that you’re not a bot. That you are real. That you’re actually, your intent is to look into the product. So you start to have a bit more skin in the game, shall I say, to move down.
Okay, so you select the top, usually the area with the highest percentage of drop-off compared to the step after and with the highest volume. As you said, it’s the first step.
Now, we know we need to work on this homepage. We’ve worked with this client. We know this is absolutely, this place, this page, we need to fix it. What do you do then? When do you start talking to customers or starting to look at session recordings or heatmaps, or whatever it is?
Barron: When I know what the issue is, and it’s where I want to spend all of my time, that’s when I dig into the qualitative.
I normally start not watching session maps or talking to customers yet. I usually just start by reading every review of the service out there today, because it is the people that are willing to be very vocal about the service.
They either love it or hate it usually, and they tell you exactly why. Either it did fill their need, and they’re so elated that they decided to go to a public forum and tell everyone about how great it is. Or they hate you, because you didn’t fill a need when you promised to.
That’s also a huge concern. You only want customers that you served their need, that are good revenue. You don’t want to trick people into using your service who would not be a good fit.
Louis: Yes, absolutely. Let’s say we are looking at a review site like G2 Crowd for the B2B sector, or I don’t know, if you work in travel it could be Trip Advisor, whatever.
Louis: You read those, and I know you have a lot of experience in that. You read those quite a lot, so you probably have a natural extinct for what you’re looking for.
But for people listening who might not have done that in the past, how do you actually digest those reviews to make sense of them? So that it’s actionable; you can take away something and do something with it?
Barron: I honestly start by reading all of them every time. I don’t start by trying to immediately cut into insights. Because for a lot of services I’ve worked on, they’ve been very different at Rev. I think starting with a beginner’s mindset every time is very helpful.
Read everything. Then you’ll start to see some common themes. If people allude to the same topic multiple times, you take a note of it.
I always have a running doc open where I put the best reviews, the worst reviews, and then, over time, the themes that I’m seeing. What are people talking about the most? What is the most frequent praise or criticism of this company, and why?
After bubbling it up into themes around what the customers speak, then you can develop ideas for, why they love it, or why they don’t love it. And how you are accurately or inaccurately addressing that on your top of funnel today.
Louis: Right, and so what do you use? You use a Google Doc or Trello board or… ?
Barron: Yeah, I just use a Google Doc. I share it with the team entirely. I think that a lot of this learning should not be done in a silo.
That’s a whole different conversation we can have. But I believe in sharing within the org as much as possible, because other people will have insights or ideas and see things that you don’t.
Louis: Yeah, and so the thing to say here is that humans are amazing at identifying patterns. We’re very good at reading conversations and starting to see patterns emerging.
Now, we’re not as a good as artificial intelligence. It does that way better than we do. But we are much better than artificial intelligence, at least as of 2019, to extract insight and to make changes and to make connection between things.
Yes, I would concur with what you said. Reading reviews, you start to build up the empathy. You start to understand and get people. As you said, after 10 or 20 reviews, it feels like people are repeating the same thing. Even if they’ve never talked to each other. Which is amazing, so you know you’re onto something.
You said the most common criticisms and why, and the most common praises and why. Those are the two main things you’re looking into, yeah?
Louis: Then you share that with the team. You spread the empathy. You make sure people care and focus on people who matter which are the customers and not investors and other people like that, right?
Barron: Laughs. Exactly.
Louis: You have this Google Doc, and then what do you do?
Barron: You try to turn those insights—and the disparity between what customers are talking about and what your landing page speaks to today—into hypotheses for how you can improve your site.
Barron: I always like grounding any changes I do in this thinking first and developing a hypothesis before just making changes to the page.
Even if you don’t have enough data to make statistically significant A/B test changes, you can see if all the changes you’re making are along one vein and one idea, and it’s not working, then you test a different idea overall.
Because a lot of early stage people that probably listen to this show, don’t have the troves of data that Facebook has. where they can snap their fingers and an A/B test is finished.
For a lot of us, we have to work off of qualitative feedback and can be searching around for things that work. We won’t know immediately if something’s working or not.
Yes, start with a hypothesis, a main idea that you have for why people are not buying today, and why they should be.
Louis: When do you start talking to customers like you said or looking at session recordings or heatmaps or something, surveys, maybe? Do you do that after? So you only look at reviews, or when does that come into the mix?
Barron: Oh, yeah, after looking at reviews. I start doing all of those in concert after reading reviews, which I think are the most vocal customers. Then I reach out to customers. Customers don’t always want to talk to you all the time, so it may take a little bit of hounding, a little bit of incentive.
And those get scheduled out and then start digging deep into any qualitative data possible. Including, heatmaps and session tracking and those kinds of things.
Louis: Let’s dive into talking to customers. I’m interested in the way you do it. And perhaps, I can share a bit of how I did it in the past. How do you reach out to them? And more importantly, once you have them, what do you like to ask them?
Barron: That is the great question. How do I like to reach out to them? I usually reach out by email.
Someone who I really respect in the growth world is Hiten Shah. He’s done a ton of writing on the subject. Usually, people are very, very open if you ask them something about their opinion of your service or ask for their help. People usually want to help.
Not a high percentage of people have time to help, but they’d like to think that they’re helpful people. Don’t say, “Give me your feedback,” or things that are very harsh. Usually, just asking someone for their opinion goes a really long way.
Once I have them on the phone, I usually like to ask first about the moment of discovery and their buyer’s journey.
What was happening in your life that led you to think that you needed a product like ours? What do you normally do when you search for products like this? How did you evaluate them? Do you remember or not? What made you, if you remember anything at all?
Usually, they’ve already used your product by now, if they’re a customer, so they’re going to be very biased to say that your company was far and away the best solution out there. Because they’re already embedded.
Take it with a grain of salt how much they say that you stand apart from the competition. But usually, they’ll tell you the types of things that were important to them when they were making that buying decision, what they’re doing with it now, how it adds value to their life, those kinds of things.
One other very fun question to ask is, how has your life changed since using our product?
Louis: Why is that?
Barron: Because it pushes them to think about how their life is actually better now. Once you know what better looks like to them, that is very valuable to then speak to people who have not even tried your product to say, “Your life can be better by X by using us.”
Louis: I suppose then you also probably record those conversations?
Barron: Only when I get-
Barron: Ask for consent, yes. Quick plug here, Rev has multiple recording apps. We have a free call recording app in the App Store. And we have a free, in-person voice recording app, so if you want to do user interviews by phone, you can use our app for free.
Barron: After asking for consent, please ask for consent. Look up your local laws, they’re different in every state.
Louis: Every country as well.
Louis: I would, I concur. A few weeks ago actually, I did this exact exercise. I talked to around 25 customers, and I asked them roughly the questions that you said.
What made you look for a solution like ours? What was the journey from there to there? I transcribed all of those discussions using Rev actually. Which is true, I did, and was super helpful.
Now, I’m curious. You have reviews. You have transcript of conversation of people who agreed to that. We can talk about maybe other qualitative type of data, but how do you start to make sense of all of that to start building hypotheses then?
Barron: As you said before, human beings are really good at pattern recognition. Everyone listening to this podcast is brilliant for listening. You all have the ability to listen, get information, and develop insights from it.
By looking across all of these different platforms, you will see common themes. I’ve never been in a case where there’s no one thing that stands out the most. One thing will stand out, and it will match with judgment.
As I said before, I fully believe in sharing ideas across the organization so other people will also have insights and hypotheses. By having that discussion with other people, one thing will stand out and become very clear that that should be your number one thing to try.
I say try, because as smart as we all are, it could be totally wrong. And it has been a number of times for me. It’s good to develop a hypothesis and a process for testing it.
Louis: Right, we’ll talk about that in the next few minutes because that’s actually super important. But for my experience as well, what happens is you see one or two insights are just overwhelming like they’re just trumping the rest.
You know, the 80:20 rule tends to happen quite a lot. 20% of the insight responsible for 80% of the issues in a sense.
You would identify themes, right? Those themes—to go back to the topic of the show—are they more related to customer concerns, or tend to be more related to potential wins or things that we need to double down on? What do you usually start to see to be the themes that relate to those?
Barron: I’ve actually seen both. I’ve seen that it’s customer concerns that you’re not proactively addressing in the first place, or there are huge wins in the product that you never talk about. At Rev, it was actually mostly the latter.
We had built up a service that was incredible, thank you for being a customer. We have a Net Promoter Score that’s off the charts, and we were not accurately telling people about our service when they came to our website.
Most of the wins came from just telling about the great work we had done, because our customers were doing a better job of selling than we were.
Louis: Yeah, that’s usually what happens, quite a lot. It’s the gap between what your customers, what your best customers think of you, and the concern they had in their head before using you. And the things they really love, and what you’re actually displaying. The closer you match those two things, the better it is, right?
Louis: All right. Now, we have a list. Let’s say we have a list of, maybe you can give us an example from Rev—you talked about the NPS, the fact that you have a Net Promoter Score that is off the chart. And you didn’t necessarily display that on the homepage or wherever else. Then how did you go about, “Okay, this is a clear insight,” to “This is what we need to test and in this order?”
Because what tends to happen a lot is people, you have a lot of people just say, “Oh, well, then it’s obvious, we need to do X.” But you have many, many different solutions to the same problem. So how do you start gathering those solutions to this problem you’re trying to solve?
Barron: Yeah, once you have the problem very well defined, I believe then that you should try to look anywhere possible for inspiration. Whether it be internally from your team and gather everyone who thinks they’re really smart and has a solution. Gather all of those, create a running list—we do it in a spreadsheet—but create a list of every single idea that someone has, and what hypothesis it addresses.
I also look externally. We are smart enough to know that we were not the first people to work on building trust or explaining the benefits of our service. So we look for other companies that did this extremely well.
Usually, you can look at very fast-growing startups likely do this very well, or products and services that you use and love today likely are very good at this. I would look for reputable brand names. Look at experts. See what other people are doing to address the same concern, because you’re likely not the first person to need to tackle this problem.
Once you have a running list, you need to come up with a framework for evaluating all of your different ideas against each other. We use a relatively simple framework at Rev. It’s called ICEE. It’s used a lot in conversion rate optimization.
You evaluate the impact of the change that you’re going to make. If it is a big change, it will likely have a huge impact. If it’s a tweak on a button color, which we don’t do at Rev, then it will likely have a very small impact.
C is the confidence that you have that it will address the hypothesis, and it will address the customer’s concern. Then, E is the effort or ease of the change. Changing a picture above the fold on a website when people first come to your site is a lot easier than redesigning the entire website.
I would figure out what is very impactful to do, what is very likely to address the customer’s concern, and you’re confident it does. And what is relatively low effort? Something quick you can do.
Normally, these take the form of text changes above the fold. Are you talking about your product, or are you speaking to how you make customers better, potential customers’ lives better?
Louis: Thanks for that. I think in particular the confidence one is something that people fail at accurately scoring.
Barron: It’s hard.
Louis: They had a high score when their CEO said, “No, no, no, it’s definitely going to work.”
A way that I found was actually helpful, I interviewed the VP of product at Doodle a few months ago. And he was saying that the way they track confidence is they do the tiniest test possible of the idea using anything like a fake door type of, a fake page just to test action of people, or they beta test it. Or they do something very, very quick to see, “Okay, that’s moving the needle.” They are quite confident now that actually the confidence is quite high or not that much.
But, yes, I think the enemy in this type of scoring is when people start to put their own preconceived idea of, “This is absolutely going to work,” into the confidence thing, because that could-
Louis: -skew the results.
Barron: Totally. And as you get more repetitions and do it more and more, you’ll get a better sense for how something will work. But we’re constantly surprised.
Another way I like to validate tests, and this can get messy if you’re running the process for the first time. But once I start developing really good ideas, hypotheses, and then have some top tests that I’m thinking about doing… and I’m still doing customer calls in parallel.
I will ask customers about it. I will ask customers what they think about different verbiage, or whether we try something. And I’ll get them to do it in real time, because they’re the people that you want to connect with the most.
Louis: Absolutely. I guess the biggest win in terms of copywriting, because you said copy, changing copy is something that works quite well. Because it’s easy to change, yet it could lead to a lot of impact, is to literally re-use what people say about your product.
Barron: Oh, yeah.
Louis: It sounds cliché, and it’s something I’ve talked a lot on the podcast. Joanna Wiebe, for example, talked about it on this podcast and a few others, “Good copywriters don’t write copy. They steal it from customers.”
It’s definitely the case when you try to come up with the smart marketing copy, and then you ask a few customers, “Would you say this one?” They say, “No, that’s sounds … That’s not clear. I don’t understand it. I wouldn’t say it myself.”
Usually, that’s kind of a first principle. The simpler something is, the more likely someone is to trust it. Remove all the jargon, keep it simple, and use what people say. That should lead to some healthy results, right?
Barron: Totally agree.
Louis: Right. So let me go back a bit in time, like five minutes ago, where you were starting to talk about collecting insights and ideas. I suspect, for one of the core themes—let’s say we select a theme that we want to tackle.
We create a spreadsheet, you share that to all your team members, and people will start to just write ideas.
An idea into something that is actually prioritized, I suspect there is a few steps in between, right? How do you turn that into an actual hypothesis from a growth perspective or a marketing perspective? I think you have a structure for that.
Barron: Yeah, I think you’re alluding to my post. But I make it as clear as possible: what the problem statement is, the change that you’re about to enact, and the expected results that you have in terms of conversion, or what you expect to happen.
While that is going on, you start all of the other steps necessary to actually get a change to happen.
If it’s a change on the website, which it usually is for top of funnel. Then if it needs design, kicking off the design process. If it requires copy changes, figuring out exactly where you’re exactly going to change copy, and what it’s going to say. All of those kinds of things.
Louis: Give me an example of a hypothesis. Just something like that, to see how it looks like in a sentence.
Barron: Yes. A good example would be the current state of our landing page does not fully address the customers’ concerns around speed or address our turnaround time for our transcription service.
If we accurately address and show customers what the turnaround time will be, then our conversion rates will improve. Because customers will know exactly when they should expect their audio file returned in a written out format.
Louis: Thanks for that.
Barron: That was on the spot, not as eloquent as I could be.
Louis: No, it’s all right. It’s good. I think people definitely understood the point.
Once you have those ideas from all the teams, and I suspect you need to process them into a real hypothesis—because sometimes it could just be a few words put into a spreadsheet, so you need to digest them and process them, yeah?
Barron: Exactly. You need to clarify as much as possible. I’d say the more you can write about your hypothesis and clarify what the issue is the better it will lead to future insights.
Then it will also help clarify what the test is going after. Whether the test works or not, you can point back to the hypothesis and say, “This was correct or not,” so being specific helps.
Not saying, “Our landing page is horrible, so no one wants to buy for us. If we make it better, everyone will buy.” That doesn’t benefit anyone.
Louis: Yeah, that’s exactly what I wanted to get to as well. You need to be specific. You need to be very clear and attach a metric, a success metric that is relevant to this, right?
So you would score them based on what you said. What happens then? Once you have, let’s say, you have a prioritized list of items, what do you do?
Barron: You start going after the first one, and you start designing your test and seeing what it would look like. Whether it needs design or writing resources, picking the top test, and starting to put it into action. What that looks like depends on the stage of company you are, and what resources you have at your disposal.
A startup, using Instapage or something where you can quickly make changes, may just throw up the new one. Because they have such limited data that anything is better than their current baseline. But if you are a more mature organization like Pinterest, you have a very set infrastructure for how you roll out changes.
Then you work with the teams appropriate for getting the designs necessary—and all of the infrastructure or functionality ready to go—then you make sure that all of your data systems are working before you implement the change.
I’ve seen people do everything in between as well.
Louis: At Rev, in particular, do you like to merge a few test ideas together into one big change— or do you like to do them one at a time?
Barron: In an ideal world, you could test every single small punctuation change on the site. And if you’re Facebook, maybe you can do that. But in the real world, A/B testing isn’t as clean as that. So, yes, we have definitely rolled together a couple ideas that test the same common theme.
Overall, it has been net positive. We have seen incredible improvements from that. However, the downside is that you’re not exactly sure what piece of the different things rolled together actually makes the difference.
Sometimes, that’s okay. When you’re working with limited data, that’s just the way the world is and you have to make that work for you. But, ideally, you could isolate changes and know their impact.
Louis: Yes. Because what could happen is, let’s say we are grouping five test ideas together. You implement all of them on the homepage. It turns out that you had an increase in conversion, which was what you were looking after.
But it’s possible that one out of the four actually brought down the conversion. Or people didn’t necessarily like it. You don’t know that and you keep pushing for it. That could have a negative impact.
As you said, if you’re a small company, you don’t have a lot of traffic. Let’s say like 200, 300, 500 visitors a day, don’t start playing with small A/B tests with copy changes because you’re not going to see any change.
Definitely do collect all the qualitative feedback, quantitative data you can get, and bundle them all together to do one big bang, which is how you call it I think. One big change, a massive update based on the best data you have, which is your customers and go for it.
Because you will never have perfect data. Even with Pinterest or Facebook never have perfect data. The closest you can get is definitely by being very close to your customers. If you have a good empathy for them, you will be able to empathize enough to create things that they will like.
Barron: Yeah, and I think that’s the essence of marketing as well.
Louis: Yes, people tend to forget that a bit. Because in marketing, there is the word market, and they forget that.
Let’s say we’ve tested, so what happens next? Once we’ve tested the few first tests, what do we do?
Barron: Well, you see whether the test worked or not based on whatever instrumentation you have in place. Hopefully by now, you’re a hero in your organization, and conversion rates have skyrocketed.
Because the test was amazing, and it accurately addressed the hypothesis. You analyze, “Why were we right? What went right about this process? What was wrong? What are the learnings that we can take to other parts of the product or other parts of the organization?” And you share those.
The flip side is if you’re wrong, you need to do the exact same thing. You can’t pull any punches and say, “Oh, whoops, we slightly had a wrong hypothesis. We’re not going to share these results across the organization.” No, you do the exact same steps whether you win or lose, because you learn something no matter what.
That’s the reality of testing. You’re never going to be right 100% of the time, usually not more than 50% of the time, so you have to know that going into it. Every single change you make, there’s something that you can learn from that. Sharing that back to the organization so everyone gets brought up to speed and benefits from that test.
Louis: How do you share those learnings specifically to people who are not really very well versed to A/B tests and growth and conversion rate optimization? How do you share that in a format that they would understand?
Barron: I do it through Slack messages where I just write out exactly what we’re working on: the problem, the conversion rate we’re trying to improve, the problem that we thought was at play, the changes that we made. Results, and then, why we were right or wrong, and what we’re doing next as a result.
Because this result will greatly inform your next steps. If you don’t use this data which is the most real data that you can to inform what you’re doing, you’re doing the process wrong. This feeds back into the initial step.
Louis: It seems like you’re putting a lot of importance in your work into clarity and simplicity, which is great, because I completely agree with you. Keeping it simple and clear for everyone. One of the reasons why everyone hates marketers is when we tend to over-complicate and use jargon. If you clarify and simplify, it gets better, yeah?
Barron: Yeah, I have the benefit of not coming from a marketing background, so I have to use simplest words because I don’t know the jargon.
Louis: Good. Yeah, I think that’s a massive advantage.
Okay, so thanks so much for going through all of this process with me. That was quite detailed and in depth, I think people enjoy that. Is there anything I forgot to ask you on this, or that you wanted to mention on this process?
Barron: It will be a living and breathing process that should likely change and evolve over time to the needs of your organization. Every company does it slightly differently, and that’s okay.
Louis: What do you think marketers should learn today that will help them in the next five, 10, 20, 50 years?
Barron: We’ve talked about this a lot already, and I think it’s true that people don’t change that quickly. I don’t think people change. So if you learn how to study and learn from people, you’ll be in a great place.
Because tools come and go, technologies change, the latest growth hack changes almost weekly, but people don’t change that quickly.
By studying and learning from customers, you cannot go wrong. As you said, market is the core of marketing, so understanding your market and what you’re trying to achieve, you will end up in a great place.
Louis: Yeah, amen to that. Maybe as a follow-up question then, what are the top three resources you would recommend our listeners maybe to understand their customer better? It could be anything by the way. It could be a book, a podcast, a video, a YouTube, a conference, whatever.
Barron: Yeah, that’s a good question. On top of all of the learning tools which are the current hot things that people are using today, I subscribe fully to Nassim Taleb’s Antifragile book where he talks about the Lindy Effect, the longer something exists, the more likely it is to continue to exist.
I look to classic marketing and advertising books to help inform decisions that I make on a very regular basis. I’m going to go with older books. One that I’ve written about is Scientific Advertising by Claude Hopkins, which is an amazing classic by this man, Mr. Hopkins, who is the father of A/B testing through direct mail.
He writes principles that I came to my own conclusion with by doing A/B testing before I found him and then realized, “It’s the same exact thing.” Nothing has changed in almost 100 years. It’s fantastic.
Another one is Influence by Robert Cialdini, who is another incredible marketer. Who talks about how people don’t change, and the principles that drive human decisions. Once you read this book, you’ll see his principles played out all throughout the world of marketing.
A final resource that I tell everyone on my teams to read, and that’s true of marketing and just in life in general is How to Win Friends and Influence People by Dale Carnegie. Again, it’s about how to connect with people, which I think is the core of marketing.
Louis: I couldn’t have said that better myself. Thanks for sharing all of those. I think people got a lot out of this episode. You went really in depth, and I’m very grateful for that, and I know if you’re listening to this podcast, you’re probably also quite grateful.
If people have questions about what you said, or if they want to praise you or insult you, where can they connect with you?
Barron: All insults can be sent to my email: email@example.com. If you want to read more and visualize a lot of the things we were talking about today, then you can look at my Medium where I’ve written about a lot of these concepts that we’ve covered.
Expect more writings in the future. And if you want me to go deeper on a topic that I haven’t already, please let me know.
Louis: Awesome. Once again, Barron, thank you so much.
Barron: Yeah, thank you, Louis. This was great.