How to Tell Better Stories With Data

Businesses try to squeeze every lead through the funnel, get you hooked on their products, and produce garbage content to capture your interest. Don’t feed the bears! And by that, we mean, as consumers, don’t give away your email address. And don’t pollute the Internet with unoriginal content. Today, I’m talking to Alexandra Samuel, a freelance writer, research, speaker, and author of Work Smarter with Social Media. She helps companies cope with the transition to a digital world. Break through the noise, and get people’s attention using the Internet.

Listen to this Episode:

How to Tell Better Stories With Data

00:00 /

Topics Discussed in this Episode:

  • Changes and competition regarding the creation of content
  • Process of telling stories with data
  • Provide original data or repurpose it to stand out from the crowd and gain attention
  • Telling a good story is not a guarantee of anything; get your story out and about
  • Step 1: Pick a topic/objective and map out a message that you are trying to deliver
  • Step 2: Start with a headline and develop questions
  • Two camps come with data storytelling: One camp just wants an interesting headline, and the other camp needs a specific type of headline
  • Vast majority of social media content comes from less than ⅓ of companies’ customers
  • Frame questions in a concrete way to get accurate and specific results
  • Data-driven projects are more of an investment than other kinds of content
  • Step 3: Sourcing the data and evaluating the angle
  • Perform Google searches to determine if topic/theme is worthwhile or overdone
  • Single best cure for marketing is to think of marketing as a service
  • Is the data useful and interesting?
  • Step 4: Collecting good data, if available
  • Fastest and cheapest way to get good data depends on levels/standards that you want
  • Search for open data, publicly available data, and other phrases to find data
  • With surveys, take it yourself and ask others to take it before deploying it
  • Finding angles/topics using your own data (Web traffic, purchase history, etc.)
  • Meaningful influencers and indicators
  • How to present data even if you are not good with graphics or mathematics
  • When it comes to design, give necessary data to designer, and determine what’s your most important takeaway
  • Dumbing down online conversations as dollars flow in that have no purpose other than driving purchases and competing for attention; the result is an explosion of crappy content
  • Alexandra hopes the Internet turns into an asset; displace toxic and terminal levels of material consumption with sustainable forms of online experience and consumption  
  • Create digital experiences that create meaning
  • As a marketer, what are you trying to accomplish with your talents for marketing?



Louis: Bonjour! Bonjour! Welcome to another episode of, the marketing podcast for marketers, founders, and tech people who are just sick of shady, aggressive marketing. I’m your host, Louis Grenier.

I don’t really remember how I came across this subject, two years ago, I stumbled upon an article called The Social Cost of Bad Online Marketing on the Harvard Business Review. This article talks about how businesses are all about lead generations and squeezing every lead possible through the funnel, how to get people hooked using the product. It’s all about producing garbage content to get people interested which leads to race to the bottom.

The author also talked about that perhaps we should stop feeding the bears as consumers, as people by giving your email away, to download a shitty ebook, or to sign up to webinar. Maybe companies would then start to produce contents, and therefore the internet will become a better place. This was one of the main inspiration to this podcast, Everyone Hates Marketers. I’m really happy to have Alexandra Samuel who’s the author of this article in this podcast today. Alexandra, thank you so much for being here.

Alexandra: Thanks for inviting me.

Louis: Let’s talk just briefly about who you are because if I read your full bio, we will basically talk for 20 minutes. I tried to summarize it as much as I can but you’ve done a lot of things in your professional and personal life. You’re a freelance writer, researcher, a mother of two, you’re a speaker, the author of Work Smarter with Social Media, a guide to managing Evernote, Twitter, LinkedIn, and your email, that was published in 2015. You also contributed to The Wall Street Journal and Harvard Business Review, as I mentioned.

You help companies cope with the transition to the digital world by tackling everything from the business and social impact of big picture trends, like the emergence of the collaborative economy, to the nitty gritty of making productive use of digital tools. I think that’s a nice summary. Obviously you’ve done much more than that and people can check you at There’s one big problem that you like to tackle and that you seem to tackle with your clients and the readers, it’s really hard to break through the noise and get people’s attention nowadays especially in the internet.

Alexandra: It is hard.

Louis: Why is that?

Alexandra: I think we have moved very quickly from a world in which, the web one world of static websites to the web 2.0, it seems […] call it that. But the web two world of user generating content has led to such an explosion of content, and of course, the development of Facebook and Twitter, and other social platforms, means that everybody is a publisher now. We’ve moved from a world in which the creation of content was relatively narrowly controlled to a world in which it’s democratized, which is wonderful, but it means that basically everybody is competing for the same number of eyeballs but against a much larger set of competitors.

Louis: This is becoming more and more of a problem. I feel that when I talk to listeners and marketers in particular, I feel that they are really overwhelmed by all of this and they really don’t know what to choose, what tactics to use, what strategy to use. They’re really drowning in a sea of possibilities. Today, what I like to go through is a subject that you are an expert in.

We talked about story telling a few times in this podcast before. We also talked about in the other episodes about making sense of the data that you have available to you, but we haven’t really talked about mixing the two. Telling stories with data, this is something that I like to go through with you in a step by step. Hopefully, marketers and listeners can take that away and apply that in the business. Why do you think telling stories with data could help marketers and listeners rise above the noise and break through the noise out there?

Alexandra: I think it’s helpful to frame this with a little bit of my own personal journey through the space. I’ve been involved in social media since before it was called social media, I guess I started blogging in 2004 when it was relatively easy as certainly by today’s standards to get attention. I didn’t have that problem that a lot of social media users have come online were recently half of just trying to even get a toe hold.

I had been blogging on my own site, blogging on some other sites. I’ve been a contributor to the Harvard Business Review for a while and I’ve had some blog posts over the years that really blew up and some that disappeared, that’s just natural for anybody who’s publishing online. And then I found myself, I guess about five or six years ago now, at the time I was running a research center at Emily Carr University here in Vancouver.

We had partnered with a company called Vision Critical which is a customer intelligent software company to do a project around data visualization. Vision Critical was trying to think about how it could make customer feedback data more compelling, more visual engaging. Emily Carr is an art and design university so we had a whole bunch of design students who are interested in the opportunity to work with original data. We did a study through Vision Critical, we did a big research survey of how people use social media.

We took a slice of that, one of the things that came out of that was about Pinterest which was just getting started at the time. One of the students who worked with the Pinterest data just created this brilliant infographic or data visualization to talk about how Pinterest was putting people in stores. I ended up turning that into an article for HBR, a blog post for HBR with an original infographic to go with it that was designed by this very talented designer named […] who’s now at Lemonly.

Like I said, I had plenty of hits at HBR over the years. But the reaction to this one post was so enormous that it was really an eye opener for me, it just spread everywhere instantly. Years later, it was still getting shared. That made me realize that in this increasingly noisy world where there’s so much content being created all the time, providing original data or even repurposing existing data in an original and visually compelling way is almost the sure fire way of standing out from the crowd.

I say that with caution because if there’s one thing I’ve learned from the art of working on the social web for more than, gosh now it’s getting close to 20 years, it’s that one day, sure fire hit. In a few years everybody is doing it, it doesn’t have the same currency. At this moment, still, sharing an original data driven story is far more likely to garner you significant attention than just about any other kind of content you can create.

Louis: In the future, as you said, the currency, the value of such an activity may decrease slightly but it’s still rooted in first principles. Telling stories will always work. If you tell a good story, people are wired to listen to good stories. Using your own data means that you have an angle, you have something that is quite unique. Therefore, it has never been shared before. The combination of the two makes sense in the long as well. Yes, things might change, but unlikely.

Alexandra: I think you’re right in a sense that good stories will always resonate if people can find them. Anyone who’s a writer or blogger will have stories like this. I’ve written some pieces that I am so proud of and I’m so in love with and spent so much time on that 40 people have read. Telling a good story is not a guarantee of anything. The problem is how do you get your good story out and about? What I found is that when you are sharing a story that is data driven, there are always gonna be these little bite size nuggets that other people wanna pick up.

Part of the value of the data driven story is you’re much more likely to get media covered, social media coverage, people sharing the infographics that you included in your report. That can drive attention and traffic back to your story.

Louis: Let’s get into methodology to help you as a listener to understand how to tell good stories with data. What would you say is the number one step? What do you usually start with with clients and with people you advice?

Alexander: What I usually start with actually is the headline. I like people to start by thinking about if we’re gonna do a project, what would be the dream headline to come out of this? It’s quite important, I think, when you’re doing a project like that to recognize there are two camps when it comes to data storytelling. In one camp, you just want an interesting headline, and the other camp, you need a certain kind of headline.

Let me give you an example. With Vision Critical, we have done a number of these large scale pilot project studies that show the companies on Vision Critical’s platform how they can use social media data and survey data in combination or what kinds of customer feedback service might help them chart a new direction for their business. When we did that, what we cared about was showing people that they could find something exciting, we cared about an exciting headline but we actually didn’t have a dog in a race.

It really didn’t matter to us ultimately whether Pinterest was putting people into stores or pulling people out of stores. What you’re looking for there is just a sexy, interesting, surprising, and ideally counter intuitive headline. In the case of one of our social media analytics studies, we showed that the vast majority of social media content, actually close to 90% of what companies see on social media from their customers, comes from less than a third of their customers. That one finding is the headline you can build a whole program around.

A lot of the time, if you are doing a data driven study, you actually can’t be that agnostic. You need to have a certain kind of headline that drives your particular business model or business value. For example, there’s a home improvement company called HomeAdvisor that did a really wonderful example of infographic data project that I often show people that showed the total amount of money that Americans lose to home improvement scams every year.

If it was a small number, it wouldn’t be a very sexy headline and it also wouldn’t really tell the kind of story they wanna tell which is that you need to be sure that you’re getting a reliable contractor. Starting from a headline, starting from what would be a finding here that would really cause a splash and serve our story, if we have a particular angle we needed to serve, and then working backwards from that.

You can’t count on getting the story you want out of a data driven project at all. But I find that starting by thinking about what the kinds of headlines would ideally be really helps to inform the way you approach projects so that you do come up with some interesting findings.

Louis: Let’s say we want to do a massive survey to all of our customers and find out what they think of marketing as a practice and how much do they trust marketers. If I named this piece like The State of Marketing in 2018, it might not resonate as well as something such as only 7% of people trust marketers and 93% of people don’t trust marketers. You’re thinking already about the final result. If you’re listening to this, you might not know but the headline is the number one thing to nail when you publish something new because this is the first thing that people would see.

Alexandra: Let’s play with that example for a minute. Let’s say you know, you wanna put out a data driven piece of content to promote this podcast. You are trying to tell the story about how everyone hates marketers. Coming back with the result that says 93% of people don’t trust marketers, that’s a good result. First of all, you have no idea if you’re gonna get that. Second of all, what’s the survey question you asked to get that? The only way you can really go about getting that data point is through a survey.

If you ask people a question like, “Do you trust marketers?” It’s kind of a meaningless question in a way. But people probably will, a high number can probably will say they don’t, but what does that really tell you? That’s where thinking in terms of more concrete kinds of questions, I guess we can think of this as our second insight here.

When you’re designing a survey, really try and ask people both a range of questions that you have some room to play with your results, maybe you don’t get the headline you thought you’re gonna get but you find something interesting instead. Also, frame your questions in a really concrete way because you’re more likely gonna get an accurate result.

For example, give them a scenario, you are thinking about purchasing a new car. You encountered the following sources of information, a television out about the car, a brochure from the dealership, a conversation with a friend who recently bought the similar car, a series of Facebook posts by people you don’t know about their experiences with the car, or a magazine review of the car. Ideally, you’re not even just saying trust, you say, which is gonna have the biggest impact for you on judging the specific characteristics whether this car has the specific features you need, what’s gonna help you decide if this car is a good value for money, what is gonna help you decide if this car suits your lifestyle. The more specific you can get, the better. Ideally, you’re even asking about a specific purchase.

Thinking about the last time you bought a car or ideally thinking about more recent, thinking about the last time you bought a new kind of cereal. What shaped your decision to buy that cereal? The more you can get people to think about specific purchase decisions, the more you can really drill down and get them to be specific about what influence them, and then think about what you end up with.

You’ve asked people about their last cereal decision and then they’ve told you the different things that factored in. I tried it in a friend’s house, my friend told me they liked it, I saw a bunch of people posting about it on social media, I saw a TV ad board. You then break up those results and say, “Here are the sources that came from marketers, here are the sources that came from peers. Look, the ones that came from marketers, nobody really listened to.” Why? Because people hate marketers.

Louis: How would you categorize this advice? I think it’s still a bit blurry in my head in terms of the type of questions you can ask because you have so much experience in that, it might sound easy for you. I’m just trying to think in my head and listeners’ head how it can even be more detailed. I think we’re missing a step, though. Yes, we’re coming up the headline, we’re coming up with the questions or the survey. But how do we even start to pick the topic or the theme? How do you typically advice people to do that in the first place?

Alexandra: It really depends. I mentioned already that you have some scenarios where you’re just looking for a headline because frankly you’re just looking for attention and inbounds and mentions. You’ve got scenarios where you’ve got a specific message to deliver. If you have a message that you’re specifically trying to deliver like your example, let’s show the people who don’t trust marketers, your data driven content project is gonna be more obvious, you know you’re gonna do something around how consumers feels about marketing.

If you just came to me and said, “I am Acme Cereal Incorporated. I just want people to hear about us, it can really be anything, we can do anything that has to do mornings or breakfast or health or food.” We’re wide open. If all I’m looking for is attention, how do I choose what my topic area is? In a sense, that’s no different from any other content marketing scenario where you’re just looking for attention and you don’t necessarily have a subject matter. You’re looking for themes that underline your brand’s value proposition that resonates with your brand’s positioning.

The only difference really is, I try and look for data driven projects in areas that are not yet colonized. If you’re Acme Cereal and you wanna do the stated driven project, survey is about people’s healthful eating. There’s so much data out there on eating habits and what food is good for you, what food isn’t bad for you, why people eat what they do, forget, you’re not even gonna get any attention. You might get attention for something quirky and funny about morning, what makes mornings difficult for people. Morning people versus night owls and what helps them get going in the morning.

You won’t always be able to do your data driven project in a wide open space particularly if you have a particular message you’re trying to deliver. But if you’re just looking for brand awareness, look for the most wide open territory you can find.

Louis: The first step is probably to map out all of the potential angles you could use. That comes back from an interview, one of the episodes with Lexi Mills where we talked about digital PR. She talked about some few very similar but finding the right angle and the right theme to start with. Straight away, you talked about cereals and you made the association with mornings, you made the association with healthy eating, you made the association with milk, whatever it is. Straight away, I can visualize a mind map of some sort where you start drawing ideas, anything that relates to your product or service.

Alexandra: The truth is it’s much harder to do this in an abstract than it is in a concrete situation. When I’m working with a specific brand and we’re having these conversations about what kind of content project they wanna tackle with data, there’s always something you’re trying to tie into, it’s the campaign themes of the year, it’s the new brand position, it’s a new audience or market you’re trying to tap into. You’re never in wide open territory.

Frankly, if you’re in wide open territory, you shouldn’t be trying to do a project like this because it’s just throwing spaghetti against the wall. A data driven project typically is more of an investment than other kinds of content so you don’t wanna do it until you have some validation that this is a useful subject or area or audience to be addressing.

Louis: I think step one is actually to have an objective in the first place, something that you want to achieve.

Alexandra: Yes.

Louis: Step two, trying to find angles, trying to find themes and topics that could potentially be of interest. Something that, as you said, funny, cracky, original, something that hasn’t been colonized, as you said. Do you have any advice on that on how to find the right angle? The headline is, as you mentioned, maybe step two. Step two will be the headline and then step three will be working backwards. How do I find something associated with that?

Alexandra: Once you have some vague area of the territory you’re working in, very quickly, if becomes a question of sourcing the data. Honestly, the sooner you can get to that conversation, the better. I can’t tell you how many times I’ve started down the road with somebody to think about the data driven project and then we’ve quickly discovered either that there are 50 other things just like this that there’s no point in shouting about the noise even if we think the 50 white papers already out there suck, that doesn’t mean we’re necessarily gonna be able to make ourselves heard.

Louis: How do you actually find out whether a space, a particular topic, is polluted to the point or colonized to the point where you can’t really make a dent?

Alexandra: It’s straight up Google. Typically, I do Google searches on breakfast foods ab survey, breakfast foods ab data. I do searches for basically whatever our theme is plus the word data or plus the word survey, I typically search on both, white paper, that kind of thing.

Louis: Let’s say you searched for cereals, mornings plus survey, whatever it is. You find 50 of them, what is the threshold? This something that you have a feeling for and it’s difficult for you to describe or do you have threshold to say, “Fuck that, there’s just too much out there.”

Alexandra: It’s very contextual and it really depends on who I’m working with and what their budget is. If I’m doing a small project with somebody for $10,000 and we’re aiming to do one feature infographic and a few blog posts, then the space needs to be more open. Whereas if I’m working with somebody who’s spending tens of thousands of dollars and planning on spending $10,000 and more on data acquisition alone, then we can be the dinosaur who comes in and stomps on all the little surveys. I have done that a few times, it’s really fun.

Louis: I think for listeners out there, some of them might work in both of the budget size that you mentioned, some of them for even smaller size. I guess the smaller the budget, the smaller the resources you have, the nicher, the deeper you have to go in one specific angle.

Alexandra: I think there are always exceptions, it really depends on your market. If you have a local market, you can do a local version of a data driven project and have the definitive data source on what car purchasing looks like in Des Moines. You can stand out that way. It’s gonna be hard. Honestly, I don’t think the world needs another data driven project on CMO spending, we got a lot out there on how CMOs use their budgets. How many more CMO service? I swear to God, I don’t know how any CMOs have time to do any of their work because it seems they spend their whole lives answering surveys about CMO spending.

Louis: That’s so true, I’ve seen so many in the past. There’s a lot like this, the state of X in X years. Do yourself a favor and don’t pollute more space that is already full of things that have been done before, instead, maybe take a bit of a risk and try to find an angle that is a bit more original.

Alexandra: Actually this ties into the theme of your podcast. I think the single best cure for bad marketing, in my view, is to sincerely think of marketing as a service. I think about a launch I had with a colleague of mine many years ago, at the time when I was doing only cost related marketing and he was doing a lot of commercial marketing. I was shocked but also touched that he was a true believer and really believed that these big commercial campaigns he was doing for a video game companies were of service because they help people know about the game.

That might not have been my jam but I really do believe that’s the lens through which we need to approach marketing if we want it to be at least not garbaging. Data driven projects are really important check on those instincts because you have to ask yourself, “Is this data gonna be actually, if not useful, at least interesting?” If you wouldn’t find this survey to be interesting or useful or if you wouldn’t find this analysis of social media data or mobile app data or spending data, whatever data you’re gonna be working with, if you aren’t personally curious about the results, then forget it, it’s not an interesting project.

That, I think, is really the place you have to check all of these projects from because they do take resources and they can be really fascinating. It almost doesn’t depend, for me, on the topic. I get super curious about a range of subjects once you start looking at real data around it but you do need to have that personal curiosity driver.

Louis: Let’s take a step back. Step one is to come up with your objective, the key things that you’re trying to achieve, step two is to come up with the headline, step three is to drill down into the type of the angle, the theme that you’d like to go for. I assume, at this stage, in step three, you’ll see lots, 100% sure, of the actual angle you’re gonna use because you need to collect data and the data will also inform the angle. You might find something that you have never would’ve thought about from the data.

Alexandra: My political science professors would speak very strongly to me if I advised anyone to think that they knew their outcome before doing their research. What I would say is, especially if you’re spending any kind of money at all, you really need to be sure that what you’re doing is something that will be interesting and publishable for you regardless of what the outcome is. You don’t wanna spend $25,000 doing a whole bunch of research that ultimately shows that your product or your brand is terrible.

You need to be approaching your data driven projects in a way where you can be a little bit agnostic about the outcome. Even if you go in with some hopes for what the headlines will be, you’re gonna have to stay open and actually let the data speak to you. You can pick and choose, this is an academic publishing here, this isn’t journalism. If you’re doing content marketing, you’re under no obligation. If you do a study of transaction data as an indicator of where people’s financial priorities are, you’re not under any obligation to share every possible finding from that.

If your findings about breakfast foods are that people are happy as if they’re spending more money on eggs than on cereal, you’re not under any obligation to publish that finding. But you’d better hope that there are some other stuff there like people who wake up earlier are happier than people who wake up later and then that’s the finding you publish on step, I hope that’s not true though.

Louis: Step four then is collecting good data, how do you go about this?

Alexandra: Step three and step four have to happen in tandems, you have some ideas about what you wanna pursue and then you have some ideas about how you’re gonna source your data. You essentially can’t walk your topic until you walk your data source, particularly if you are not in a position to collect original data. When it comes to thinking about data sources, you are basically doing one of three things–you’re either repurposing public domain or possibly licensing preexisting data, you’re basically working with somebody else’s data.

You are using data that you have that’s proprietary to you that you already have in hand, like your customer purchase data or social media, different businesses have access to different kinds of data sets. You may have data at your fingertips already and I find that’s often the case with many, many companies. They have data already but they haven’t been thinking of it as content, they’ve been thinking of it as customer intelligence, it’s like, “Let’s look at the data we already have, we’re already looking at our web traffic, we’re already collecting data on our customers transactions. How could we turn that into content?”

That’s the other piece, shopping from your own warehouse of data. Of course the third option is to develop an original data set, deploy a survey, setup some social media monitoring and use that as a data source. There are other ways you can go about getting data, those are probably two of the most common. That will almost always be your most expensive option, is having to collect new data. But bear in mind that there is a bit of tradeoff between collection cost and analysis cost.

You may already have an incredible amount of data on purchasing habits and a whole bunch of loyalty card data to go with it, but depending on how that data is structured and organized, and also what regulatory concerns there are on how you repurpose it. It may be very hard to work with whereas for $1000, you can run a survey of 1000 people or 2000 people and it would be much more easy to analyze.

When you’re thinking about what your data source is gonna be, there’s 5A which is think about what your options are, but then there’s 5B which is think about costs. When you think about cost, don’t just think about the cost of acquisition, think about the cost of analysis.

Louis: From your perspective, what is the fastest, cheapest way to get good data?

Alexandra: I think this is such a dodge but good data is a very contextual definition. If I’m developing a cancer drug and I wanna know about side effects, my standard of good data is very high. If I’m doing a funny infographic for Acme Cereal on what people like to do in the mornings, then good data doesn’t have to be very good at all.

Louis: Let’s take an example then, I’m picturing those people listening who might not have a business themselves or maybe in a small businesses and they’re looking to try something like this. I will assume that a mix of data that they already have with their customers plus maybe a survey might actually be the easiest option. I think buying all the people’s data might be very, very expensive and that might lead to too much cost. What are the examples of angles you’ve used in the past and perhaps you can draw some conclusions from the data that you have yourself. You mentioned purchase history, purchase behavior, that kind of thing, traffic, you even mentioned. What are the kind of angles you worked on in the past on those type of data?

Alexandra: I also wanna mention that not all preexisting data sources cost money, there are huge amounts of data available in the public domain. I would really encourage people to go out there and take a look at what’s available. I did a piece for JSTOR Daily a few months ago, there was this great controversial article about the impact of mobile phones on teens’ mental health. I found tons of publicly available data online that allowed me to do a new analysis. Don’t feel like just because you don’t have a big budget, you can’t do this kind of work. There are lots of data out there. Sometimes you can do things, they’re very clever with what’s already there.

Louis: How do you find it? How do you find this data?

Alexandra: I’m not gonna dictate URLs over the recording here. If you basically do a search on open data, publicly accessible data, there are a number of university data collections that are available, things like the general social survey. Often, it’s really academic research because in academia, people do tend to make their data sets open and often are required to do so. There are lots and lots of data sources out there that you can find, and some that are pay per use but not necessarily very expensive.

By the way, another one is in terms of creating new data sets, I’ve become totally obsessed with Amazon Mechanical Turk because you can do a decent survey of 500 or 1000 people on Mechanical Turk for a few hundred dollars. The Mechanical Turk surveys aren’t perfect but if you read up a little bit on methodological issues, it’s actually no worse than any other data source.

Louis: Alternatives to Mechanical Turk, SurveyMonkey does that, Google actually does that, I don’t remember the name of the service but Google allows you to survey panels of people. There are a lot of ways to survey randomers.

Alexandra: The most important thing I would just say for people who are doing surveys is take your survey yourself and then ask other people to take it for you before you deploy it because you often don’t realize the problems with your survey questions or how unanswerable they are until somebody else takes them.

Louis: I asked you the question and then we went on a tangent as usual, I tend to do that quite a lot. How do you come up with angles on your own data? You said web traffic, purchase history.

Alexandra: I think we’re back to that original question of what’s your topic, that really depends on whether you’re going for general brand awareness or specific message and thinking about headlines. What I would say is think of some possible topic areas and then it’s like, “Here are 5 ideas for potential 5 or 10 or 20 ideas for potential data driven stories, here are the 5 sources of data we already have access to.” How do they match up? Where is their match?

Louis: Do you have any examples in the past of the type of data like this that you managed to use?

Alexandra: Last year I worked with Sprinklr which is a social media platform that has quite a breadth of data. They partnered with Forbes to develop the list of the world’s 50 most influential CMOs, I was the data journalist on that project. It was really, really interesting because Forbes had done versions of this report previously. But because Sprinklr has such an extensive set of social media analytics and because LinkedIn was a partner and then they had a set of social media data, of LinkedIn data.

We are able to work with Sprinklr who just acquired a company called Little Bird that has really cool tools for visualizing influence networks. It was really cool actually because it challenged me to think about influence in terms of both the different dimensions of influence but then also how that manifests online, what are the things that we wanna look for in order to assess influence meaningfully. Ultimately, we made a point of trying to have a pretty eclectic range of indicators. It wasn’t just social media mentions and it wasn’t just social media follows but it was also mainstreaming the mentions.

It was also the Little Bird data that looked at the inner connections with other influencers and the LinkedIn data that showed connectivity within LinkedIn. By combining all of those dimensions, we were able to come up with a much more robust measure of CMO influence. Also, to complement the court list of here are the world’s 50 most influential CMOs with a bunch of other insights that surfaced about big picture patterns in CMO influence that surfaced through the process of looking at the data and making notice.

Louis: We’ve been talking about this for a while now, we haven’t talked about the visual aspect too much because this is also about what you do best. You’re able to identify stories, find the right data in order to tell this story. But then you mentioned infographics and things like this in the last few minutes. How do you advice people to present once they have this data in front of them and to those result in front of them?

Alexandra: There are actually two really important related things. One is I am the least visual person on the planet. The other is that, until I started working with Visual Critical and finding my inner data nerd, I could not even do long division anymore. I am not in numbers, I do not think of myself as a math person, I do not think of myself as a visual person at all, I think of myself as a storyteller. I think that’s really important.

I think people often get scared away from doing data driven content because they feel like they’re not good at making infographics or they’re not good at numbers, these are all manageable problems. When it comes to visual piece, a few different piece of advice popped to mind.

If you’re working on a big dollar project, you want a dedicated information designer to work with you on your infographics. That’s a real luxury but it makes an incredible difference. When you’re doing that, you wanna make sure you’re working not just with a graphic designer whose work you like and whose aesthetic you like, but with somebody who has a specific aptitude and experience around information visualization.

One of the reasons I’ve gone back to Sheryll again and again on many projects is that she has the ability to come up with infographic designs that are not only clever and engaging, but that are able to capture multiple dimensions in a single image. That’s really powerful. If you have the privilege of working with a dedicated designer on this kind of project, two pieces of advice. One is think about what your most important takeaway is. I mentioned that insight from our project on social media analytics around how the vast majority of social media content comes from very small number of users. That was our most important takeaway.

That was where we really encouraged Sheryll to go to town, do something super cool. She ended up making this boombox that visualized the concept of instead of noise versus signal in social media. I’d rather see people spend half their budget on one graphic and the other half on the other eight so that that one graphic really stands out and gets picked up, gets shared, and so on.

Second piece of advice is when you’re working with a designer, make their work easy for them. I always create an Excel workbook with a separate sheet for each infographic I need, each chart I need created. I put just the data I need, nothing that won’t appear or inform the graphic, and I put at the top of each sheet here’s the key message, this piece in intended to convey, this is what this graphic needs to show, and then I create a crappy version of the graphic in whatever Excel will do and say, “This shouldn’t be a pie chart, it doesn’t need to look pie chart but here is what I’m trying to show.”

You make the work of the designer much easier. I’ve had many designers tell me that that’s super helpful. That’s all if you’ve got the budget. If this is more of a DIY thing, then what I often find myself doing is just trying to get a step up from Excel. If I’m working in Tableau for my data analysis, sometimes I create my images directly in Tableau. A lot of the time I use a service called Infogram which basically just creates the same range of charts that you would get from Excel, you just take your Excel Spreadsheet and paste it in there but they look a lot nicer and you have more control over the visualization, I’ve done a lot that way.

I also have become obsessed with Creative Market which is a peer to peer marketplace where designers sell different kinds of creative assets. I have purchased assets that have allowed me to make my charts look nicer, my infographics look nicer. When I was doing this big project around digital parenting, I created this infographic called What Kind of Digital Parent are You? It was like one of those long infographics and then I had a bunch of charts within it.

I want your people to know, I am not a designer, I cannot draw at all but I’m really proud of this graphic because I was able to take the charts, I created an infogram, and then restyled them using assets I bought on Creative Market, and complement those illustrations I got on Creative Market. I ended up with a really unusual looking and distinctive infographic that looked, I think, like it have been designed. I’m sure professional designers will tell me they could tell it wasn’t done by a professional designer but I’m totally proud of it. I think those are all tools that can really be helpful for people who are trying to DIY it.

Louis: Thank you so much for going through all of the range and the possibilities scenarios. As you said, people have money, people who don’t necessarily have the resources to do that. Regardless, I think, a lot of people listening would have designers in their team. Whether they’re specializing in this or not, as you said, if you can prepare the ground for them and really make it easy for them to discover the data and really make it easy for them to understand what is the core information you want to display, then there’s a good chance that this will turn into a good piece.

I think we’ve gone through this extensive step by step together. Thank you so much for doing this, I know it’s not that easy to go through all of those steps in an interview like this. I’m not a visual person either so I’m gonna check your infographic and see if it’s good, no I’m not. You thought originally that we would be talking about your article about The Social Cost of Online Marketing but I’m really a bad online marketing and really thought that this topic was even more practical and it was something that really people could take away.

I don’t know if we have time to go in depth around the topic of this article that you mentioned but let’s give it a chance for the next few minutes. I tried to summarize this article at the start, in the intro, but perhaps you have a better way to explain what’s going on by online marketing, what’s happening there? Perhaps give us a glimpse of the future of internet as it is right now. Where do you see it to go in 10 years on even 20 years?

Alexandra: I need to be honest that I approached this conversation like the proverbial, cranky old man saying get off my porch. I have been working on internet stuff since, wow it’s been more than 20 years no. I did my first serious internet research projects started in 1997, I did a research when Bob Putnam was writing Bowling alone, the book, he wanted to look at the impact of the internet on social capital. I was in grad school at the time, I was his research assistant to do that research and that was more than 20 years ago now.

I’ve been tracking the social impact of the internet, this is horrific. It’s basically half the time that it’s been around, that’s such a weird thing to realize. When I did the research for Bob, I was like, “The internet is gonna be collate for social capital.” I had a totally different data than what he did. In 1997, when I set out to do the research on the internet’s impact on social capital, I literally read every single data that have been written about the social impact of the internet because there’s wasn’t that much, there were like 50 or 100 articles at the time.

I really held fast to that optimism. It’s quite funny, at every stage I’ve been like, “No, social media is gonna be great, mobile is gonna great.” I always wanna believe that it’s all gonna be great and then I get gradually dissolution and depressed. When I first got involved in social media, I didn’t even think of it as marketing and I probably would’ve done it if I’ve known it was gonna be marketing. I think partly because I think of marketing as selling stuff. I think that fuelling people’s desire to acquire more stuff is the problem, not the solution.

When I got into what you would now call social media marketing, it was before Twitter and Facebook, my husband and I started social signal in 2005 and said, “We’re gonna start this web agency and we’re only gonna do Web 2.0 projects.” I think we were literally the first people in the world to say we’re only doing Web 2.0. We started by building online communities for change oriented, like social change organization.

That was pretty much our whole practice for five years, we basically just build online communities for different flavors of save the world kinds of organizations. We worked with the elders, we worked with tech soups/complement for ties. Here in Canada we worked with VanCity, we do a lot of really cool projects. It was such a privilege to be involved in the social web at that time where we really thought like we were fostering connections and conversations among people who really wanted to change the world and really wanted to use social media as a way of deepening connection, sharing ideas to solve problems.

I was optimistic in those early days because I was like, “Coke is gonna spend money but who wants to talk about Coke?” At the end of the day, it turns out that if you are prepared to dump enough dollars on the table and make really cool videos, you can get people to talk about Coke. I should say I’m talking about the beverage and not the drug. They may not care about Coca Cola as much as they care about climate change but if climate change has depressing videos and Coca Cola has really cool videos, that’s where the conversation is gonna go.

What I’ve seen over the years has really been this dumbing down and coarsening of the online conversation as dollars flow in that have no purpose other than driving purchases, and then of course competing for attention of people who could be watching people jumping out of hot air balloons, you have every other small company just trying to get attention. You have this explosion like, “Crap content.” I just cannot believe how bad the vast majority of content marketing is.

I’m not talking about bad strategy or lousy web coding, I’m just talking about this 500 word article has no purpose, says nothing and enlightens me in no way. Of course there’s the endless spam, the endless online advertising. We really are in this arms race for attention that is leading people to create more and more crappier and crappier content, and driving out the actual sources of good content in the form of traditional investigative reporting.

Louis: Where do you think the internet is gonna be in 20 years?

Alexandra: Where do I think it will be or where do I have a small hope of it going?

Louis: Let’s keep it to the optimistic view.

Alexandra: In all seriousness, this is my approach to this. Honestly, I’m really pessimistic but what’s the point? If we’re all gonna die in a ball of fire, there’s no real point in preparing unless you’re prepared to go offgrid in a serious way. There’s not really any point in putting energy towards that, I’d rather focus my energy on the small chance that we can turn this incredible technology into something that could actually be an asset to us as a species and improve our life and our condition.

I would say the number one way that I hope the internet can support us going forward is by displacing our really toxic, I would argue, terminal levels of material consumption with much more sustainable forms of online experience and consumption. Let’s just be concrete. Air travel is gonna get more expensive buying some major breakthrough in energy sources. We’re gonna look at a world where it becomes less and less accessible to go and have that European adventure or Latin American or whatever it is.

But AR and VR are gonna get better and better. Maybe we’re gonna have really fulfilling experiences in VR that allow us to taste some of the experiences we would otherwise have to consume huge amounts of carbon to have. Maybe instead of going out and buying a whole new wardrobe, we’re using on demand services to get just the pieces we need. I think the more that our digital lives can shift us away from consumption and towards borrowing and experiences, the more we lower our footprint.

Seriously, I don’t need to see any more cute cats, I just don’t, I don’t need any more cute cats, I don’t need any more that’s such a cool thing you just bought, I don’t need unboxing videos, enough with the commodity part. Let’s talk about creating digital experiences that actually help people create meaning.

Louis: That’s quite an answer, thanks for making me think. I was starting to get a bit weird thinking about these VR and AR stuff, travelling to Vietnam without actually travelling to Vietnam and just living this experience digitally from the comfort of your home because in the future we’re unlikely to have jobs as well, at least not jobs as we think today. I don’t wanna talk about it, let’s move on. What do you think marketers should learn today that will help them in the next 10 years, 20 years, or 50 years?

Alexandra: I’m gonna reframe that slightly and say I think if you’re a marketer, what do I say to the 20 something marketers who I’ve worked with. I love that, actually, I love the energy of some of these folks who really have grown up on the internet in a way that I obviously didn’t. I think what you really wanna think about as a marketer whether you’re early in your career or midcareer or late career, is what am I really here for? What am I trying to accomplish with this talent I have for marketing?

It’s a funny thing to think about because my 11 year old son has decided he wants to be a marketer, God help us. He’s quite funny, he’s autistic but he’s also very, very verbal. He’s very rigid like a lot of autistic kids and something like pitches on things, he’ll do such a sales drop to try and get a new video game. This is gonna be a really great experience, it’s gonna bring us closer together, blah, blah, blah, and he’s got the whole pitch.

I’m really trying to redirect that energy to be like how can you use your talents to get people to take actions, they’re gonna actually make the world a better place? Just this morning when he was giving us a pitch for some video game thing rather than talking about this marketing aspirations, we said, “How would you use your marketing talents to raise money for autism?” That doesn’t mean that everybody who’s in marketing should be going to work for Greenpeace or Unisoft, but I think you should ask yourself.

As a marketer, my job isn’t just to sell the stuff other people create, I’m creating value. Increasingly, marketers are creators as well. That’s the whole shift towards digital and content marketing, we are creating content all the time. Ask yourselves, what can I create that has meaning? How can I build on my own skills and talents and passions so that as much of my career as possible is really almost expressive and using my passion to bring something beautiful into the world?

If you’re a young marketer who’s also a really passionate cook, go and work for craft and think about how to do recipe and food content that bring people together around what you love about food. What is it that makes food special to you? Why are you such a passionate cook? Think about what makes that meaningful to you and then find yourself a marketing career that lets you express meaning. I just come back to that again and again. The more marketers can focus on making meaning, the more we can have a positive impact and also careers that we love.

Louis: This is why this interview is getting longer than an hour, it’s because I completely agree with you and this is such an interesting subject. By the way, one of my younger brother has autism as well and he has Asperger’s Syndrome, he’s verbal as well, and he would be very much like you described. He’s not into marketing, he’s into visiting one supermarket every Saturday and spending eight hours a day, and spending time talking to his favorite shop assistant and all of that. He’s working full time but not as a marketer, thank God. I definitely connect with you. What are the top three resources you would recommend to listeners?

Alexandra: I’m going to resist the urge to self promote so just scratch that and I’m just gonna go full nerd and talk about my favorite digital tools. First of all, how much have you guys already talked about Canva? I still evangelize Canva to people a lot but if you’re thinking about creating shareables, I just go back to it again and again, I love it. Another tool that I have recently become really fond of is Streak which is a CRM that integrates with Gmail.

I’m just like me, freelance writers. Thank God, I don’t have a sales pipeline that I maintain per se. I use Streak to manage my story cue and I have a workflow. I actually just wrote a piece for Zapier a couple months ago about my workflow for capturing my story ideas and moving on to Excel and then started moving on to Google Sheets, now I just tweaked that. Instead of landing in Google Sheets, it all lands in Streak. I just think, for a huge range of marketing tasks, if you’re a Gmail user, you’re using Google hosted email, Streak is a really fabulous tool.

And then, because I like to give you a wild card that I bet nobody else has mentioned, my favorite thing that I own on the internet, I have a bit of a domain name habit, my favorite thing is my URL shortener, which is If you go to, you’ll find my blog post about Evernote. If you go to, you’ll find about my advice about being a digital mentor parent. If you go to, you’ll find my speaking page.

When I first decided that I’d have a Vanity URL Shortener, I looked at Bitly, I looked at the other options. But the promise if you host the Vanity URL Shortener with Bitly, if somebody else has got a link that points to bitly/speaking, you can’t use /speaking. You end up having to use whatever random characters Bitly assigns to your short link or a really long alternate.

I use a self-hosted tool called YOURLS. That allows me to own my own namespace so I can have whatever I want as Plus, this is where it’s really handy, unlike Bitly, when you create a Bitly link, you can never change where it points to because otherwise people would be spamming. But with YOURLS, you can do that, this is actually part of my data storytelling strategy.

When I create a report that has infographics in it, I create an short link to each graphic and then I can go back in once I’ve uploaded all of those images to Pinterest, I can make the short link point to itself so I can post something on Pinterest that says I can go and make sure graphic one points back to that graphic that I’ve already uploaded, that’s a bit of a nerd point. Bottomline is you gotta have your own URL shortener.

Louis: You’re quite a nerd, I have to admit. I didn’t think you’re that much of a nerd but now you’re really a nerd. Alex, you’ve been an absolute pleasure. Thank you so much for going through all of those tricky questions that I’ve asked you. Where can listeners connect with you and learn more from you?

Alexandra: I’m on Twitter as awsamuel although I have to admit, I’m not as avid as a Twitter user as I used to be. You’ll probably find more of me on Facebook, facebook/awsamuel. My main home is my own website which is

Louis: Alexandra, once again, thank you very much for your time.

Alexandra: Lovely talking with you.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.