Paul Boag Shares a Favorite Ecommerce Project

In Part 1 of my interview with Paul Boag, we discussed the importance of making a name for yourself in the web industry. Content marketing has played a key part in his own success and that of the agency he co-founded, Headscape. Of course, getting the “gig” is only the first hurdle – you have to deliver good work too. In this excerpt, I ask Paul about his experiences working on ecommerce projects. What follows is an illuminating and unusual insight into an ecommerce site catering for a demographic where the typical user is over 60.

In Part 1 of my interview with Paul Boag, we discussed the importance of making a name for yourself in the web industry. Content marketing has played a key part in his own success and that of the agency he co-founded, Headscape.

Of course, getting the “gig” is only the first hurdle – you have to deliver good work too. In Part II of my interview with Paul Boag, I ask Paul about his experiences working on ecommerce projects. What follows is an illuminating and unusual insight into an ecommerce site catering for a demographic where the typical user is over 60.

You might also like: Partner Spotlight: Jivaldi Builds American Sniper Store

Keir: You’ve worked on a few ecommerce projects over the years. One that you have spoken about before, Wiltshire Farm Foods, sounds like it posed a number of interesting challenges due to its niche audience.

Paul: Hugely! Wiltshire Farm Foods was a website aimed at elderly people. It sold frozen ready meals to them. It's basically a paid version of “Meals on Wheels.” The average customer was in their 80s, which was incredible. I give a talk about this at a conference once and I put up a picture of two old people. What I say is one looks slightly older than the other. I said, "That one on the left, the younger looking one, that's the son." He's in his 70s or 60s.

It was a crazy audience to work with, but not necessarily a bad audience. You'd think you'd have all kinds of problems and certainly there were some issues – mainly related to accessibility. People of that age suffer a lot from arthritis, so we used to wear ski gloves to try and move the mouse and click on links. It gives you a sense of what it's like not to have full motion control using a mouse. The other great thing I would do is take off my glasses and try and navigate the site, because I'm as blind as a bat.

"We wore ski gloves to move the mouse. It gives you a sense of what it's like not to have full motion." Click to tweet

In other ways, they were a great audience. Because they lacked confidence, they would read every piece of text on the page – which meant they actually read instructions. They are the only audience in the world that actually reads your copy.

Keir: What kind of things specifically did you have to focus on?

We concentrated on removing a lot of clutter. Wiltshire Farm Foods had all these different categories of food, but in reality, customers only ever bought from the top eight categories. So we hid away categories that people didn't really look at much, just so they weren't feeling overwhelmed.

Another thing we did was provided lots of visual feedback, as these users needed lots of reassurance. For example, when customers add something to their shopping basket, normally you would just update the number, wouldn't you? That was nowhere near enough for these people.

Another thing we did was provided lots of visual feedback, as these users needed lots of reassurance. For example, when customers add something to their shopping basket, normally you would just update the number, wouldn't you? That was nowhere near enough for these people. The product would actually fly across the page. It sounds ridiculous, but it would animate across the page into the shopping basket. Then the product photo would then have a highlighted border around it. The button which did say, "Add to basket" now says, "Add another to the basket." You had a little tag that said "1 in basket" and the basket itself would obviously update. All these little touches added to reassure the users.

When it came to completing the information about yourself and your delivery address, we put a little tick by every field as they moved from one field to the next that said, “You're doing it right.”

Then we spent quite a lot of time thinking about objection handling, which is a sales term that means working out why someone might not order. With this user base, it was external factors such as having a strange man turning up at their door. To help with this, we started to explore areas beyond the website. One thing we did was get the driver's CRN checked in order to reassure people. We also made it so that the driver would actually come into their house and stock up their freezer, as many users had real mobility problems.

We also didn't ignore the fact that this audience wasn't particularly comfortable online. So when people entered their post code, we would show them a picture of their franchise person, and give their name and telephone number. It humanized the person that was going to be delivering to them, and allowed them to pick up the phone.

"We didn't ignore the fact that this audience wasn't particularly comfortable online." [Click to Tweet]

We also allowed people to reorder at the door. Now, from an ecommerce point of view that's a really bad thing to do as you're losing track of the sale. You no longer know how effective the website's being because the user is ordering at the door and you don't know they were an ecommerce customer. It was kind of bad for ecommerce, but right for the user and right for the audience. We did a lot of that stuff; looking at those kinds of bigger picture issues.

We looked at really basic stuff as well, such as ensuring that buttons looked like buttons and that links had underlines on them.

Keir: I know you had some fun testing the site.

Paul: The first time we did usability testing, we got a few users into a usability test lab. We sat them down with a desktop computer and a nice big monitor. We thought they'd need a big monitor, and a mouse, and a keyboard.

The first little lady came in and sat down in front of this computer. We said, "The first task is..." and she said, "No, dear, I can't do that."  And we said, "Why not? What's stopping you?" "Well, I don't know how to use the mouse." I'm like, “What?” "No, I have a laptop with a touchpad on it." We kind of dismissed that.

The next person comes in, sits down, and we tell them the task. I swear this is true. They picked up the mouse and put it on the screen. And so it went on. Not one of the six people we tested had ever used a mouse. They'd all only ever been given laptops. They'd skipped the whole mouse stage.

None of them knew how to do it, so we kind of gave up on that round of user testing. Instead, we decided to go into peoples' homes and actually test with their real kit. It was great, an amazing experience. I spent 40 minutes in a user-testing session, an hour and a half having tea and biscuits hearing about the Battle of Britain. It was wonderful.

We decided to go into peoples' homes and actually test with their real kit. It was great, an amazing experience. I spent 40 minutes in a user-testing session, an hour and a half having tea and biscuits hearing about the Battle of Britain. It was wonderful.

But the best thing was this one lady. One of my questions was, "Okay, I want you to add a product to your basket." She did that. I was about to say, “Now I want you to go to your basket,” and I thought she wasn’t going to be able to do this. It was because she had a Post-It note stuck over her monitor in the top right-hand corner, which was where the basket was. She wasn't going to see it. Another lady we tried to do testing with had a cat climbing over her keyboard most of the session.

It was a hugely enlightening experience. We learned loads from that, but the main way that we discovered all this stuff was though relentless A/B testing, and I mean relentless. It was obsessional. We managed to get, at one point, a 6% jump in conversion rate by changing one line of text. The website had a ridiculous conversion rate anyway. We got the conversion rate up to something like 32%, which is phenomenal isn't it?

"We managed to get a 6% jump in conversion rate by changing one line of text." [Click to Tweet]

Over the five years we worked with them, we increased their sales by 10,000 percent on the website. The 6% jump we got was because of the Verisign logo. You know the Verisign logo people used to put on their websites? We had that at the bottom. We want people to know this website is secure, so we put the Verisign logo on. “This website is protected by Verisign.”

I was wandering around the site one day, clicking through things as you do. I thought, I bet our audience doesn't understand that. “Verisign, what the hell is that?” So I thought, let's do a bit of A/B testing. We tested a variety of different versions, one of which was just a padlock with the text, "We look after your credit card details" or something equally vague. And it caused a 6% jump in conversion rate. People were terrified about entering their credit card information, and apparently just promising to be good with it was enough to keep them happy.

Keir: That's a great story. Was the site built on a particular platform or did you develop it in-house?

Paul: It was bespoke. The main reason was that things like Shopify weren’t around when we started working on it. The site also had a lot of very specific needs. For example, one of the problems with the site was that the business operates on a franchise model. That means each franchise is allowed to set their own prices, which meant when a user arrives at the website, they can't see prices because you don't know where they are located.

When we first started looking at the previous site before we worked on it, the first thing you encountered when you arrived was an ask to enter your post code. You couldn't see any products; couldn't do anything. It was a screen with “Enter your post code” on it. There were all kinds of integration and backend issues to support the franchise system, which basically meant that it had to be a custom build.

Keir: By playing the post code lottery you could get the same goods cheaper?

Paul: Absolutely. One of the things the website had on it was, “We offer free delivery.” Users aren't stupid. They would go, “I'm being charged more if I'm in this postcode compared to that one. Therefore, delivery is not free. You're lying to me.” One user said, "I would never purchase from an organization that lied to me like that."

What we ended up doing to get around the problem (which sounds like a weird thing to do) is that we showed the highest price. Then what we did is gave people a discount when they hit the checkout. We said, "We'd like to give you a discount." We don't say why. We just say we're going to give you a discount. People love that. They think that's the best thing ever. They were happy to pay the higher price, and here we are giving them a discount.

Keir: It's been brilliant, Paul. Thank you so much.

Paul: No problem at all.