Drive and Convert (Ep. 030): Data at Each Stage of your Company

Listen to this episode:

About This Episode:

In the Ecommerce world, it’s almost always about the data. But what works well for large organizations may not work well for small ones and vice versa. A lot of that has to do with how much data you have, and large organizations tend to have more. So where should different size organizations starts as far as collecting data and making use of it? And why do we even need data in the Ecommerce world?

• Apple Podcasts
• Google Podcasts
• Spotify
• Castbox
• Castro
• Overcast
• Pocket Casts
• Stitcher
• Subscribe via RSS

Episode Transcript:

Announcer:
You’re listening to Drive and Convert, a podcast about helping online brands to build a better e-commerce growth engine, with Jon MacDonald and Ryan Garrow.

Ryan:
All right Jon, so in the e-commerce world and the world that we live in it’s almost always about the data. The more you have generally the better off you are in our world and the better you can monetize that data, even better. The problem ends up being I think as you’re going through the entire scope of e-commerce businesses, certain things that work for very large organizations are going to be very different for smaller organizations and vice versa and a lot of that has to do with how much data do you have? Because in small businesses you generally have less and large businesses, generally more. Whether they use it or not is a different story but the complex thing that I come across often is how do I, or how do we, help these small businesses take the steps with their data to become big businesses, because if they don’t do anything with their data they’re probably not going to grow unless they get super lucky, but where can they be starting in this chain and what do different size businesses need as far as data monitoring or data collection or making use of all this quote unquote data that we have available to us in the e-commerce world? At the end of the day why do we even need, I guess why do we need the data?

Jon:
Yeah. All great questions.

Ryan:
I guess that first question is probably a good one to start with. Why do we even need data in the e-commerce world?

Jon:
I think you’re right. It’s not uncommon that I see that sites have way too much data and they’re just not doing anything with it. It’s like you keep piling the dirt on the mound but you never actually build anything. Right? I think it’s interesting, vast majority of people would say you need data to make data-backed decisions. But is that it? No. I think that the real goal here is to help brands get closer to their customers. It’s to help them get outside that jar. I always talk about reading the label from inside the jar. It’s near impossible. Data is one way to help you get outside the jar, and it’s obvious it’s also to help understand what’s more obvious to their visitors. A challenge that your visitors know about and have is not going to be as obvious to you as it is to them in a lot of cases.
It’s to help you see outside the jar and I think a lot further down the list in the reasons why you should be having data and tracking clicks and movements on your site is to make data-backed decisions. Yes it’s on the list because that is important and I do believe, look, I run an optimization firm, we make data-backed decisions for customers, but it’s not the main reason or even one of the top reasons. I think that that’s important to point out.

Ryan:
Got it, and when you’re saying data, because data could mean lots of things, what specifically from your perspective is data?

Jon:
Yes. User engagement data for the most part. What we want to do is understand how people are using something, why they’re taking those actions, and what they’re thinking. Often I say you want to track every click and movement that’s happening on your site. Yes, but that only tells you what people are doing. It doesn’t tell you what they’re thinking or why they’re taking those actions.
I generally look at this as four main areas, and even within those four there’s quantitative and qualitative. Quantitative is all about the numbers. All the data that goes into the actual numbers. Analytics is a great example of that. Then you have the qualitative side that is more customer interviews, talking with people where you’re not getting data so much as you’re getting an understanding.
Those four areas I would generally look at, one is analytics. That’s really important. Every company has analytics installed. If you don’t, I don’t know what you’re doing but go set up free Google Analytics and get it on there. Your eyes will be opened to the world. The second is engagement data, and I’m talking about how people are using your site. What they’re doing on your site. This is all the things like heat maps, click maps, scroll maps, eye tracking. You can do session recordings. All of that data is how people are using your site. What are they engaging with in terms of the content, how far down the page are they scrolling, what are they clicking on, what are they looking at, what order are they doing things on a site and what are they going through the funnel? All of that type of engagement data is the second one.
The third one is user testing. Now this is where you would send people to your site who match your ideal customer profiles. You’re going to ask them to complete some tasks while they’re on your site, and what you want to do from there is have them do something we call The Good think out loud protocol. That’s all about having them tell you what they’re thinking as they go through these tasks. Now it’s really generic, high level tasks that you want to tell them to do. You don’t want to say, “Go find this specific product on the site.” Now that might be a good example if you’re asking somebody to test your way finding or your search or something of that sort, but if you are really just looking at what are the high level tasks people are trying to do on your site and having a problem with and what are they thinking as they go through that, you want high level questions to start with.
Okay, so user testing will tell you what people are thinking as they’re doing these tasks on your site and engaging. The fourth is A/B or multivariate testing. You need everything else above on that list. You have to have the user testing, the engagement data and the analytics in order to be able to do A/B testing appropriately. That’s why it’s more of a graduate level course. A lot of small brands shouldn’t really even be thinking about that yet, but the goal here is to really form a good hypothesis and without a good hypothesis there is no good A/B test. I would argue and some would argue there’s luck involved, in which a lot of cases happens where small brands launch A/B tests and they get lucky and find something that works really well. That’s great, but it’s not a scientific process when you do it that way and you really want to, A/B testing needs to be a scientific process to be most effective. You have to have all the other types of data first.
Generally I think brands should be paying attention to those four types of data. Analytics, engagement data, user testing and then A/B or multivariate testing.

Ryan:
Almost generally each one gets a little more complex through that process and so at least we can start with a baseline of okay, if you’re not doing this you can’t generally do that one, and then you can’t generally do the one after that.

Jon:
Yeah for the most part. My only thing would be user testing, you could do early on and I highly recommend, I think I’ve talked about just taking a laptop and going to Starbucks.

Ryan:
The coffee, I was going to say that.

Jon:
But here’s the thing, how do you know what questions to ask? Unless you have some understanding of where people are dropping off in the funnel or what they’re engaging with and more importantly a lot of times not engaging with, you don’t know where to start with that user testing.

Ryan:
Mm-hmm (affirmative).

Jon:
That’s why I put that third on the list because if you know your analytics and you’ve done some engagement data then you have an understanding of how people are using your site, you can start to dive into why they’re using it in that way.

Ryan:
For me being in the e-commerce space I understand conceptually what each one of these are, but despite all my experience in it I probably don’t necessarily know where do you even start with getting some of that data onto your site? I would be worried to a degree about all right, if I put heat map, eye tracking, session recording stuff on my site does it have a slowdown effect on my site and actually hurt other things on my site? Is that something you come up against when you’re talking to business owners about this stuff?

Jon:
It’s a concern I hear about quite often. I think that it’s not without merit, but it’s given too much weight. Now a big reason I say that is when it becomes a problem, I talked earlier about having this mound of data that you don’t do anything with, that’s when it becomes a problem. So many companies have all these tracking scripts on their site that they just load into Google Tag Manager, if they’re lucky, because then it helps make that less of a burden on the site load, but a lot of times they just add them to their header with small JavaScript snippets and the site gets passed around between contract developers or different e-comm managers, nobody knows what anything’s for, next thing you know they’re loading 30 or 40 scripts and they’re like, “What do these all do?” Now they’re afraid to remove anything because they don’t want to break their site.
One of the first things we end up doing often when we work with especially these, I usually see this in mid-sized clients, is we do an audit of what they have installed and help them understand what each script is for and what they need and what they don’t need. That alone, I think when you reach that point, yes it’ll slow down your site because you’re loading so much stuff. But if you’re doing this correctly, like almost every site on the internet has Google Analytics installed. Google’s not going to penalize you for having their own Analytics installed because they actually get more data out of that so it’s good for them.
Same thing with you use something like Google Optimize to do simple A/B testing or segmenting of your traffic, et cetera. That can be great and it’s really not going to slow your site down very much. There’s tools like Hotjar where you can do all this heat mapping. The nice thing about these tools like Hotjar is they do a lot of these things with one script so you’re not loading a bunch of different tools.
Now there’s a bunch of tools out there that will do heat mapping but most of them will only do heat mapping. They won’t do session recording, they won’t do et cetera, so the reason that we like Hotjar at The Good is because it does six or eight different things in one script. You really just need to be smart about this. Also you really need to audit this every quarter. Look at your site and understand what you have installed and then turn off and remove the ones that you don’t have installed and if you want to get to the next level, spend a little bit of time learning it, use Google Tag Manager because it makes that so much easier to audit what you have installed. If you direct everything through Tag Manager, not only will you in a lot of cases speed up your site because it will load one script instead of 40, you will also be able to do a lot of cool things by sending data between the platforms. There’s a lot of great things you can do there. As well, also it will make it a lot easier in the future to add scripts if you want and just have a good central repository of what you have on your site.

Ryan:
And Google Tag Manager doesn’t cost anything, right?

Jon:
Right.

Ryan:
Okay, so there’s no reason you shouldn’t be leveraging that wonderful tool no matter what size business you are.

Jon:
Yep.

Ryan:
Google Analytics, Google Tag Manager, keep it simple.

Jon:
Yeah. Google’s giving you the tools. They are flat out giving you for free Analytics, Tag Manager and Optimize, the three main things that you need. There’s rumors that Optimize is going to be coming out with heat mapping and some other stuff in it as well, in which case it would even be more reason to move on to that.

Announcer:
You’re listening to Drive and Convert, a podcast focused on e-commerce growth. Your hosts are Jon MacDonald, founder of The Good, a conversion rate optimization agency that works with e-commerce brands to help convert more of their visitors into buyers, and Ryan Garrow of Logical Position, a digital marketing agency offering pay per click management, search engine optimization, and website design services to brands of all sizes. If you find this podcast helpful, please help us out by leaving a review on Apple Podcast and sharing it with a friend or colleague. Thank you.

Ryan:
You’ve got access to all these tools, four main buckets of data, how do you decide what size business I’m at? Is it just hey, for example, Joyful Dirt does a lot on Amazon, very little on our site, we’re maybe bigger than some small businesses but our site traffic is little. What filter do you use to decide small, medium, large business for data purposes?

Jon:
I look at traffic or sessions. That is going to be the best indicator. A lot of brands start by thinking about revenue, because that’s an easy way to compare perhaps. Here’s the reality though. All of these types of data are predicated on you having enough traffic to get meaningful data back. You really need to be thinking about how much traffic you have and base how much tooling and how far down that data stack you’re going to go based on your traffic.

Ryan:
Got it, okay. What are those buckets of traffic, what do they look like for you and are there certain pieces of traffic that are more or less valuable once you get into analytics?

Jon:
Oh for sure. I usually bucket this into three groups. Like a small site, a medium and an enterprise. There’s a big gap between medium and enterprise, but that’s on purpose.

Ryan:
Okay.

Jon:
Let’s talk about this. Small would be probably under 50,000 monthly sessions. Usually 50,000 is that tipping point where the data really becomes valuable. Not that it’s not valuable ahead of time, it’s that now you can really start making some business decisions based off that data on a regular quick turn basis.

Ryan:
Got it.

Jon:
If you’re in the small, under 50,000, you want to be doing things like consumer interviews. Just go talk to your customers. That’s not on your site necessarily, but it could be through a chat tool on your site, something of that sort, but you really just want to get out into the community and talk to your consumers. Find them on social. Find them through Ads. Run an ad and say, “Hey I want to talk to you. I’ll give you $5 or a gift card for a Starbucks,” or whatever, if you’re willing to spend a few minutes with me. There’s a lot of ways to source those people.
Analytics will give you traffic trends and things of that sort, but that’s really what’s valuable at this point, is trending. More than anything else you really want to be looking at trends in your analytics when you’re in the small size because any given day you’re going to see fluctuations all over the place and you’re not going to understand why. Very, very unlikely you’re going to understand what happened in a given day. You could have gotten a back link someplace. Some blog that gets a million visitors a day linked to you and you have no understanding of how that happened until a few days later. Really focus on the trends, like month over month, quarter over quarter, more than anything else.
The other thing you can do here is user testing. You could do that but I would do it in a small more intimate setting. Not using a tool set. Maybe over Zoom, something of that sort, and you still want to follow that think out loud protocol I talked about, but if you focus on consumer interviews, user testing and then looking at some engagement data in your analytics for trends, then you’ll be off to the races there in a great spot.

Ryan:
On those consumer interviews, as a small business what are the maybe one or two simple questions you can ask them to do so you’re not getting too complicated where you may not get enough value? What are the simple ones you would start with?

Jon:
Generally where we would start is asking people to find the right product for them. Let’s just set the context here. You’re looking for fertilizer and somebody says, “Okay great. I understand that. Okay now go find the best fertilizer for your needs on this site. Go to Joyful Dirt, find the best fertilizer for your needs.” Then they’re going to be able to go.
Now obviously that’s predicated on you finding the right people to fit. They have to have a garden, have some need for fertilizer. They have to know and speak the language a little bit, but that stuff’s actually easy to find. I would not go to your customer list, but I would go to people who you target through ads or Facebook groups or even on Reddit. You can find these people real easy. Go to a gardening section Reddit and just say, “Hey, you know what? I’m doing some research on a new product. I’d love for anybody who has a home garden that uses fertilizer and is interested in doing something organic and safe. I want to ask you questions about a new product.” Get them on Zoom and just give them a high level question like that.

Ryan:
That’s great. Yeah, Reddit is a treasure trove of people that are willing to engage and help I think in a lot of ways. It’s got a lot of press around GameStop recently but you can go in there and there’s very engaged people, so I think that’s a phenomenal idea that I hadn’t considered.

Jon:
Yeah, I continually told everybody going through the whole GameStop saga, which I didn’t participate in for good or worse, but I will say that the big winner out of that was Reddit.

Ryan:
Mm-hmm (affirmative).

Jon:
It wasn’t these trading platforms, it wasn’t the people who made all the money. It was Reddit because that became a household name overnight. It was already a huge platform, but now my dad is asking me, “What’s Reddit? Do you know anything about Reddit?” I’m like, “Go check it out. You like to play chess. There’s a huge sub-Reddit forum on Reddit all about chess. Go check it out.”
I think a lot of people are finding this and especially I think a lot of people who have been involved in the internet for a very long time and who have maybe love that type of forum experience because it is kind of a dated forum experience, right?

Ryan:
Yeah.

Jon:
I mean, they’ve added some cool stuff on top of it but it’s an old school format and it’s still chugging along great.

Ryan:
It’s like Craigslist almost. Like they don’t update the platform but it functions.

Jon:
There you go.

Ryan:
Okay so we’ve established a foundation, small businesses under 50,000 sessions a month. This is what you’re going to play with and this is where you’re going to be. You take the step, you’ve grown your business over 50,000, so you’re doing 100,000, maybe that’s in your case a medium-sized business. What does that look like from a data perspective now?

Jon:
Well I think the analytics become a lot more meaningful on a short time frame. You could start looking on a weekly basis, even a daily basis you’re going to start learning things a lot more quickly. You’ll be able to see where traffic came from. Not from Google searches and stuff, but if somebody links to you you’ll know a lot quicker.
I think the engagement data becomes something that you can and should react to. All the heat mapping, click mapping, scroll mapping, all of that kind of stuff is really going to be helpful because you’ll have enough of a sample size now on every page.
The thing a lot of people don’t realize is the homepage is going to get your most traffic in almost every instance. Your checkout confirmation screen is going to get the least traffic. There’s all these steps in between the funnel and it is a funnel, so every step down that funnel you get less and less traffic to work with so you get less and less data. That’s where it’s like hey, you put heat map on your homepage, almost everybody will get valuable information out of that. But if you really want to have valuable information about the entire funnel you have to have over 50,000 visitors and so that’s where the engagement data becomes useful, is in this medium size.
User testing really can be used here more correlate with that quantitative data set. You can look at your analytics and now look at where people are dropping off in the funnel in high volumes and say okay, there’s a problem most likely on that step in the funnel. Now you could run user testing on that step of the funnel where you’re now getting a little more specific with your questions. You’re saying, “Hey, I want you to add something to the cart and then take the next step you feel is most valuable.” Right? Or the correct next step. “While you’re doing that, tell me what you’re thinking, what challenges you’re having.” You’re going to learn so much in doing that.
Then over 50,000 visitors is when A/B testing really can become valuable and possible because you’re going to get enough data back from that testing to be able to make some informed decisions, know with statistical significance what is going to be valuable and really be able to make some informed decisions.

Ryan:
Got it, so generally the same tool sets but you’re going to be able to get more data in a much smaller period of time and actually do a lot more with it.

Jon:
Yeah you’re going to be able to iterate a lot quicker and make quicker decisions at that point. By this point you’re already spending to drive traffic but this is where a lot of people really will start dumping more money into spending on traffic generation because now it becomes valuable for them. They get a payout just in addition to the sale. Obviously the sale’s what they want but they start to learn a lot by spending money on that traffic.

Ryan:
Next step is enterprise for you. Where does that threshold come into play?

Jon:
I want to say like you have about a million sessions a month. It jumps up quite a bit.

Ryan:
Mm-hmm (affirmative).

Jon:
Now the reason there’s a big gap there between that 50,000 per month and a million, which is a huge gap, it’s because there’s really not, some may argue, but there’s really not a lot of value in doing anything else until you get to that level. When you get to that million dollar level, what I’m talking about value in doing is that’s when it makes sense to have an in-house team helping you optimize. That’s when it makes sense to have a user testing platform of your own that you pay for or that you’re sourcing your own user testing at a high volume or that you’re running a lot of A/B tests and you’re being systematic about it, working your way through the entire funnel, not just sporadically testing around the site where you think the biggest opportunities are.
Also I see most brands over a million sessions a month, they’re doing all of this already. They’re likely doing this but often I’m surprised as well. We start working with these customers to augment their team or to help them take that next step into something they should have been doing and I’m shocked to see that they’ve never run a user test or that they’ve never done A/B testing or that their Analytics is a complete mess but they just have a product that’s a runaway success and so they’ve just been trying to play catch up on all this stuff the whole time.

Ryan:
Yeah, there are a lot of companies in that boat right there.

Jon:
Unfortunately.

Ryan:
Okay so obviously enterprise we know there’s things that get overlooked but the small businesses, where would you say most people are overlooking pieces of data that they could easily find the most value out of if they started doing it tomorrow?

Jon:
If you don’t have Analytics, get that set up.

Ryan:
Yep.

Jon:
Almost everyone has it, but here’s the thing. Are they paying attention to it? That’s the big issue I see, is so many brands have it installed on their site but they’ve done nothing to customize it, they’ve done nothing to send reports to the key team members. There’s some great features in Analytics where you can have it email you a PDF report every morning and that gives you a really quick in your inbox, pushed to you, glance of key dashboards. But if you don’t set up and customize your dashboards it’s kind of like trying to build a house with a hammer versus an electric nail gun. You can do it, it’s going to take you a heck of a lot longer. You really want to get in and customize Google Analytics.
Then other things that get often overlooked are people going in and they immediately jump into A/B testing when they shouldn’t have done that yet. I think that’s a big concern. A lot of folks want to start A/B testing because they hear of all the value and benefits of doing it, but I’m here to tell you as somebody who sells A/B testing as a service, it’s not a good fit for everybody. You really want to be in that medium size, over 50,000 visitors, or sessions I should say, per month level before it’s going to matter a whole lot.

Ryan:
And your Checkout button color is not going to double your conversion rate, so just get that out of your head now. Don’t use that as your first A/B test.

Jon:
It’s funny you say that because as I was saying earlier, some people might say luck is involved in A/B testing. Well, I’m pretty sure if you’re testing button colors you might get lucky but that’s going to be about it.

Ryan:
That’s funny. Okay so anything else we need to know about what people need to be getting as far as data and tools to collect or based on the size that they are?

Jon:
Look, I think that getting Analytics certified if you’re small should be the first thing you do. It’s super easy. It’s built for the store owner, not for a consultant. There are a lot of consultants who are certified but it’s something like on our team if you don’t have it, we make every single person on their way in get it and I’m talking like the interns that come in. The first thing we have them do the first week is go get that certification. It will just help you to not only get off to the right spot working with a contractor or if you grow to that level, but if not it will give you a lot more information to help you make informed decisions. That’s something I would really highly recommend.

Ryan:
That was a good one. I should probably make sure I’m still certified. Good point. Take a note. Analytics certification, get that done.

Jon:
I’ll quiz you next episode.

Ryan:
Perfect. All right, thank you Jon. I appreciate the time.

Jon:
All right, thanks Ryan.

Announcer:
Thanks for listening to Drive and Convert with Jon MacDonald and Ryan Garrow. To keep up to date with new episodes you can subscribe at DriveandConvert.com.