Assumptions are good for your content strategy


Featured speakers


Andy Healey: Thank you, friends. Can I call you friends, my content comrades? Last year I moved from England to Canada, that’s Canada. When I first moved I was so excited, I was running around like a little puppy dog. But I soon found out that everything wasn’t quite as straightforward as I thought. [audience laughs] God love it.

Simple things like trying to order food; back in England we have this meal called cheesy chips and gravy, and when I tried to order it Montreal, people were rude to me. They’d shout this French insult, “Poutine!” Turns out that poutine is the Canadian word for cheesy chips and gravy. And, get this, it’s their national dish. A country with worse food than England, amazing.

So, I borrowed some money off a friend to buy some of this so-called poutine. And when I’d gotten myself a Canadian bank account, I asked how I could transfer the money back to him. You know, using e-transfer, or direct debit, or whatever. And he said, “Oh. Just Interac it.” And I was like, “What’s an Interac?” So he told me how Canadians transfer money to each other. They’ve got this really complicated system where they send emails with secret questions and secret passwords like, “What was the name of your first pet?” And then he called me up to find out what the name of my first pet was.

So cheesy chips, poutine, and Interac—these are the basics of life in Canada. [audience laughs] And these labels, poutine and Interac, they made sense to Canadians, so they assumed that they would make sense to me. But they didn’t because I had no context. And in content, we make these kinds of assumptions all the time. We assume that our developers are following our content guidelines. This is a true story, it really happened to me. We assume that our content is useful and clear. It’s not a true story it’s okay, stand down, stand down. And, we make assumptions about people’s working environments. Cat and mouse. And assumptions like these can be dangerous, and they can cause projects to fail. But maybe these are the exceptions that give the others a bad reputation.

We’re always told that assumptions are bad, but we all believe in assuming positive intent, right? In reality, we make thousands of assumptions every day, and David talked about this yesterday. Assumptions are an integral part of our decision-making process. Really simple things like when we turn on a tap, we assume that water’s going to come out of it. Most of what we decide to do every day is based on the assumption that we’ll still be alive at the end of the day. We’re looking good. People tell me that I make too many assumptions, they don’t tell me but I know that’s what they’re thinking.

So, at a Shopify event last year, Cynthia Savard Saucier who co-wrote the book Tragic Design, and Alëna Iouguina gave a talk on the science behind decision-making. Super-interesting stuff about how the brain works to make decisions. And they concluded that ... what did they conclude … that a decision, even a bad one, is still way better than no decision at all.

And as content people we often need to make fast decisions. And it can be a good thing to be opinionated. But maybe this means making a few assumptions along the way. Is there a right way and a wrong way to make these assumptions? Today, I’m going to tell you how to make them the right way, so that you can make the right decisions for your users. We’re going to do this in three entertaining, educational, and practical steps, what more could you ask for?

Number one, it’s okay to make assumptions. Number two, make inclusive assumptions. And number three, check your assumptions. As tradition dictates, we’ll start with number one.

Back when I was at university, two or three years ago, one of my majors was philosophy. And like lots of other things I experienced at university, philosophy was quite confusing. I’d like you just to take a quick moment to think about what philosophy means for you, while I have a dramatic drink of water. Now, maybe you’re imagining the greatest minds in human evolution, building on each other’s work, striving to reach an ethical and social nirvana. You’d be wrong. Lots of you may know of Descartes, and his famous, “Cogito, ergo sum.” “I think, therefore I am.” If you Google, “I think therefore I am is wrong” there are over 200 million results. A lot of philosophy is just these really smart people point-scoring and trolling each other. People love to prove the smart person wrong, welcome to my world.

In the 1800s, a group of American philosophers decided to try something different. And no, despite the big facial hair, they weren’t trying to be the world’s first hipsters. Philosophy’s a very serious business. They said, “We can’t keep questioning everything forever. Let’s make some basic assumptions so that we can move forwards.” They took the word pragmatism from Greek, which means “to make things done.” And it’s pretty much the same as that text slogan, “Get S-H-1-T done.” I can’t say the S-word, in England it’s a really bad word and my mum might see a recording of this one day, so, S-H-1-T. These pragmatists, they set themselves some principles so that they could baseline or benchmark their assumptions.

There is no universal truth since everyone experiences their own world and environments in their own way, at their own time, in a variety of situations. If something seems to work satisfactorily, then assume that it does work satisfactorily. And always allow for the possibility that a more effective solution will be found. And if these principles seem familiar, it’s because they directly align with resources that we have as content people. User personas and journey mapping, data and testing, design crits and UX research. The pragmatists had their principles to support their decision-making, and in content we have these resources.

If you were to say, “I think this content will work because my opinions are great.” Then, that’s obviously dangerous. If you say, “I think this content will work because it fits in with previous journey mapping that we’ve done, we’ve got positive data on similar content, and it aligns with our research thinking.” Then that’s some strong supporting evidence, and a fair assumption to make. So, be pragmatic and baseline using these resources, and it can be okay to use content assumptions to get shit done, to get S-H-1-T done. Sorry mum.

Okay, number two, make inclusive assumptions. Are you ready for a context switch? You’re content people, of course you are. Back in the UK we have mountain rescue teams who are groups of volunteers that help people who are lost or hurt in the mountains. Like all good project teams, at the end of every incident they do a retro and make it available online. Some of these are quite funny. Couple stuck in a tent. I don’t know how that happened, there’s a zip.

Some of them aren’t so funny. In 2007, Jennifer and Christopher Parratt drove from Oxford up to the mountains of North Wales. They described themselves as experienced hikers. When they got there they bought a map, a compass, and a guide book, and set off to hike up a mountain called Tryfan. The guidebook said the route they were taking was easy, but there is no easy way up Tryfan. In the last 30 years, 17 people have died on Tryfan. The couple got lost and Christopher Parratt slipped and fell to his death. Now, the coroner’s inquest was unusual in that the guidebook they bought was specifically criticized. The title, Tryfan The Easy Way, is extremely misleading, the map was virtually impossible to follow.

I’ve got a copy of the book here. [shows book] Can you all see this at the back? No? Yep, yep, good. Maybe this’ll help. Now right, it’s just a big wall of text, tiny little map. It’s clear that the author made a series of assumptions. The first of those was just one word, “easy.” Without that one word, the couple probably wouldn’t have even gone up Tryfan in the first place. And then he just kept piling the assumptions on top of each other, each one making the situation more dangerous. He assumed their experience level, he assumed that they would find and read all of his warnings, and he assumed that the description and map were clear. And what’s interesting for us is that all of these assumptions could’ve been made safer by being more intentional about how content was created.

User needs analysis to identify the problem that needs to be solved. Information architecture, maybe using Dan Brown’s lenses, to make sure the content is findable at the right time. Naming patterns to avoid confusion or ambiguity, and user testing to make sure the content is performing as expected. All of these help to make assumptions safer. Like a guidebook writer, our role in content is to look after our users, to take them on a journey, and guide them through safely.

Now, compare this online guide, which makes safe, inclusive assumptions. The readers might be beginners, so it doesn’t say that it’s easy. In fact, it specifically says it can be dangerous. It gives a detailed map and it gives step-by-step instructions with photos. The writer understands that for each of his users, it’s the same route, but a different journey. And this is really important, he keeps the content clear and simple.

At Shopify, we write for the average US reading age, which is 12 to 13 years-old, and there’s a really good reason why we do this. Shopify merchants are busy people who may be running their online business in addition to having a full-time job, managing their family life, and doing a million other things. They’re also located all over the world, have varying levels of literacy, and some may not speak English as their first language. By using plain language we’re making a safe assumption. Or as Sarah Richards so beautifully put it yesterday, “It’s not dumbing down, it’s ...”

Audience: “Opening up.”

Andy: Yes. So, when you’re designing content, be focused on making safe and inclusive assumptions. Design for everyone, and don’t leave anyone behind.

Okay, number three, check your assumptions. A content story. I recently worked on a project where we made an assumption, and we didn’t think it was a biggie. We’d improved one of our features, and we intended to turn it on for all of our customers. And, we knew that lots of them wanted it, but we also knew that there were a few who didn’t. So, a week before the launch, we sent out an email to everyone giving them the option to opt-out. We assumed that they would read this email. Problem was, 40% of the emails bounced and only 22% of them were opened. This set a few alarm bells ringing.

This makes me homesick for England, it’s kind of like an average night out there. It’s lovely, it’s lovely, honestly. Lots of our customers didn’t know that we were turning this on for them. As one team member commented, “That’s not ideal.” But it was a valuable learning moment. In design and tech we like to have fun. Now, in the Shopify offices we’ve got a big slide down to the kitchens, we’ve got a music room—I can’t even play the guitar—I’m holding out my hopes that we’re going to get a bowling alley next or maybe a rock-climbing gym or something. But we are all adults, and we’re responsible for the decisions that we make. And as a team, we were responsible for our decisions and how they impacted our users, and maybe we should have done something to challenge ourselves sooner. I’m going to share with you a great tool for doing just that. Introducing the assumption slam.

So, this is an exercise that we do at Shopify, it was introduced by one of our researchers and it helps us to challenge our foundational beliefs about a project. A wise person—not me—once said, “The hardest assumption to challenge is the one you don’t even know you’re making.” And an assumption slam helps you to uncover these hidden assumptions.

So how it works, you get your project team into a room and you brainstorm all the assumptions that you think you’re making about the project. So not just content people, developers, designers, product managers, everyone in the room. You write them all down on Post-its like how you would do in a design sprint or a retro. Common assumptions could be things like, “I believe my users have a need to complete this task faster.” Or, “I believe these needs can be met with a great big red button that says faster on it.” Spend about 15 minutes doing this, have everyone individually writing down their own assumptions they think the team is making on sticky notes. And when you’re done, put them all up on the wall. And you might find that you’re making a few more assumptions than you thought you were.

So to make it easier to manage, you can group and categorize them into themes. When you’ve got a manageable set of assumptions you can turn to your assumption grid. Which, being the organized content strategists you are, you prepared earlier. Now this is divided up into four quadrants based on risk and knowledge. Take all of your assumptions and, as a team, place them where you think they should go on the grid, based on how much you know about them and how risky they are. So, low risk would mean nothing too much can go wrong. High risk could affect your relationships with your customers. Really high risk, it could affect their relationships with their customers. And when you’re done you should hopefully end up with something similar to this, hopefully for you without so many high risk unknowns.

Now you can prioritize them. The low risk knowns, you can basically ignore, they’re distractions. The high risk knowns, you should include those in your project planning, and there’s also the question, if you know this much about them are they even assumptions? The low risk unknowns, these aren’t critical but time allowing, try to understand more about them. And the high risk unknowns, these are the spicy ones, these need to be evaluated. Your aim here is to move them to the left. The left? Your left, my left. The more you know about them the less risky they become. So, how do you do this?

Begin by developing a research hypothesis. For example, if our users have a big red button they will be able to complete this task faster. Then go out and test your designs on users, analyze the results. Is your hypothesis true or is it false? And then you can knowledgeably iterate on your content and designs. Now, I can see a few of you thinking, “This sounds really straightforward, it sounds good. But, does it work in the real world?” At Shopify we have this product that I work on and it helps shoppers to skip the checkouts. So if you’ve signed up for it then when you’re shopping online, instead of having to enter your address details or payment details, you’re texted a six digit code to your phone and you just enter that. It’s really fast, 11/10 would recommend.

But texting isn’t perfect, there can be network issues, people may have concerns around security, and so on. So we designed a new no-SMS flow, and yes, we ran an assumption slam to see if we were taking anything for granted with our approach. Turns out we were making some pretty risky assumptions. Are we giving our users a clear value prop? Do they feel secure enough? Will the lack of friction freak them out? After the slam we went back to work and did more content and design explorations. We shared them at design crits, we shared them with real people out there in the real world, we became more knowledgeable. And when we were confident that no assumption was left unturned, we launched.

Now, launching something is always exciting or as we say in England, scary. Shopify pay is used on millions of checkouts, so we kept an eye on how it was performing, and it worked, our users really liked it. One of those users was our chief operating officer, and he really liked it. Now this success wasn’t just down to the assumption slam, don’t get me wrong I’m not saying that. But the slam was one of the tools that were able to use to help us build an inclusive, successful product.

Okay, so we’ve seen that there’s right ways and wrong ways to make assumptions. Next time you’re starting a project, remember that it is okay to trust your gut in your content strategy, as long as you take the time to check your assumptions, and to make them safe and inclusive using all the tools and resources at your disposal. Everything I’ve talked about today has just been made available, hopefully by someone, has just been published and made available on Shopify UX, I’m seeing a nodding head, it has, wow. Has just been made available on Shopify UX on Medium. And there’s a deck on there as well, or a link to a deck on there, that you can download and use as a template for running your own assumption slams.

And before I make way for the fabulous Margo Stern, who incidentally, her mum is in the audience tonight as well, so hi to Margo Stern’s mum wherever you are. I just want to tell you one last thing, sorry Margo. Last summer I went out for dinner and I ordered poutine because, as you know, I love poutine, it’s amazing. It turns out that poutine comes in lots of different flavors, and some of them are really spicy. I don’t like spicy, this was not a good experience. Like the pragmatist said, “There is no universal truth, even with poutine.” Thank you very much.

Join the mailing list

Sign up and be the first to know about all things Confab. We promise to email you only if it’s important.

Thanks! You have been added to the Confab mailing list.
Oops! Something went wrong while submitting the form.