This post originally appeared on Growth Everywhere, a marketing and business growth blog.Hi everyone, today's interview is with Casey Winters, the mastermind behind leading growth at Pinterest, a visual bookmarking tool which is absolutely on fire. Though Pinterest can't share lots of numbers, they're estimated to have over 72 million users, and have secured over $1.3 billion in funding. Beyond that, they've got over one billion boards with more than 50 billion pins.
Besides his work at Pinterest, Casey was also the first marketer hired by GrubHub, which he grew from three to 600+ cities, making it the dominant national player in online food delivery.
Throughout the interview, Casey shares some of the best SEO tips and strategies we've heard to support drastic growth, so this interview is definitely worth a listen.
A Growth-Focused Career Starts With Classifieds and Ends With a Social Media Giant
Casey's been working in online growth for a little over 10 years. He started working in online classifieds with a company called Classified Ventures, which was also busy starting a lot of new companies from within.
After learning about all the initiatives that grow an internet-based business like paid search, organic search, and email marketing, Casey started working directly with them.
After Classified Ventures, Casey went to GrubHub as their very first marketing hire. They were already in three cities with $1 million in Series A funding, but they needed someone to help them scale to a national level.
There, Casey focused on growing the consumer side of the business through organic search, paid search, email, conversion optimization, loyalty programs, TV ads, and so on.
He helped them grow from three to 600 cities. They also acquired a bunch of other companies along the way, and after going public in April 2014, they're now valued at over $2 billion in public markets.
After that success, he decided he wanted a change in weather from cold Chicago to sunny California, and ended up at Pinterest. There, he leads the growth team of engineers, designers, product managers and analysts.
Today, they're especially focused on international growth. Pinterest is already a 40% international service, but they really want to be majority international.
Growth Trends That Carry Over From Startup to Startup
When you have a lot of content and pages of content, says Casey, search can be a really strong way to grow a business.
But the building block of creating an organic search strategy, says Casey, is figuring out how to organize your content in the same way people search for it, so if they land on your page, it boosts their experience. And to scale, you need to be able to do it without hand-curating every single page.
For example, GrubHub was all about creating a page that would tell people everything they needed to know, show them restaurants that delivered to their neighborhood, how to order, and so on. Then, they needed to find a way to scale it that would work for any neighborhood, zip code, college campus, or city someone might be in.
A big part of it was building a back-end system that let them see when they had enough restaurants in a given location to turn on a page, or enough Chinese food in a certain city to turn on a page specifically for that.
Differentiating One Online Listing Service From Another: The Race of Innovation
When GrubHub started, it wasn't focused on online ordering as much as it was simply showing you all the restaurants that would deliver to your neighborhood.
As time went on, online delivery became the stronger value proposition they needed, so that adapted that service offering.
But after a while, their competitors caught on, so they had to adopt a new, stronger value proposition of customer service.
To do that, they promised that GrubHub would take ownership of the order no matter what problem occurred: if the food was late, the wrong item arrived, it didn't taste good, etc.
With most startups, says Casey, if you're doing something that works and are growing faster than others, your competitors will respond.
If you start winning, he says, everyone else will just adopt your tactics. So it just becomes one big race of innovation.
Building Up GrubHub's Restaurant Inventory
To get restaurants on board to grow GrubHub's national reach, they used a sales team and used a replicable city launch process.
For example, when they decided they wanted to launch in LA, they would send a couple of sales reps to talk to restaurants for a week, and then they would have 50 restaurants to start with that were ready to work with them. (Often times, they would start focusing on the most popular neighborhoods first, and then expand from there.)
Then, they would put all the restaurants into their backend system and create landing pages for anyone looking to order from those restaurants.
Then, they'd go to the press in that market, and ask the press to link to their SEO pages so they could get better rankings.
They'd keep sales people in those markets to keep growing and growing the service. But as national awareness grew, they found they didn't need to send people directly to the market to collect restaurants, they could cold-call restaurants over the phone instead… but it took three years to get to that point.
Casey also attributes the move from in-person restaurant sales to cold calling to the fact that after three years, the cities you're launching to start to get smaller and smaller
Growing Pinterest to a Massive Scale
Casey says that growth from one million to 10 or 50 million still requires the same tools in your tool kit and the same type of work as growing from 1,000 to 10,000… it's just that the scale graphs look very different.
For example: SEO.
Pinterest has lots of pins and boards that individuals take the time to curate as the best content on the web, so their content is really high quality.
The challenge, though, was that they were images, so by default, Google didn't have any idea what the pages were about. This took a lot of work with users and the data they were able to generate to really explain that one certain page was about woodworking, for example. But once they figured it out, it really helped them grow.
He also still uses email marketing.
He says it's very effective with bringing people back by showing them new content, as long as you make sure all the content is very personalized, because people on Pinterest have widely diverse interests.
Challenges in Growing SEO for a Site With 50 Billion Pages
The thing is, search engines will not crawl all 50 billion pins or 1 billion boards, so Casey and his colleagues had to figure out which pins and boards were worth showing to search engines.
To get better at this, particularly with their level of scale, they had to build an SEO experiment framework.
Because they have so much content, in order to understand if the changes they were making were helping or hurting the understanding search engines had of their pages, they have to split things out by page level and do experiments that way.
For example, they'd make a change on 10% of the boards (rather than on user profiles or pins) to see if that particular change enabled search engines to send more traffic or not.
The Specs of this A/B SEO Framework Pinterest Put Together
First, says Casey, you have to identify what pages you can group together based on their similarities. (Like pins, boards, etc.)
Then you need to create a mechanism to randomly sample a group of boards or pins, and be careful to note that not all boards and pins are created equal.
Once you've got your sample group, make a change that a search engine will crawl in the enabled group, but not see in the control group, and then think about which metrics you're aiming to improve with the experiment, and run the experiment for at least three weeks.
Casey says it's actually not a supremely daunting task, but if you're only getting 100 visits from Google per day, it might be a waste of time. This kind of framework becomes helpful once you get into the millions.
Challenges of Going International
In 2014, Pinterest was translated into over 20 languages. But since then, they've found that simple translation is in no way sufficient for international growth.
For example, Pinterest is a platform that's all about content that gets surfaced.
So the early adapters in other countries who used it before the translations were using it in English, and posting all of their content in English as well.
So, for example, even though Pinterest has been translated into Japanese, Japanese users would still be shown a ton of English content, even though the majority of the Japanese population couldn't read it.
For this, they had to reconfigure how their discovery engine worked to prioritize local content and train users that if they are pinning in their country, they shouldn't feel a need to write things in English.
This is an on-going process that doesn't scale everywhere, so they pretty much have to make the effort on a country-by-country basis.
SEO is also very different internationally.
There's lots of well-ranking content in the US, but Pinterest abroad doesn't get credit for that because it's not in the local language. So to get discovered by Google, they had to create some really detailed organization systems so Google found the local content rather than the English content, which was also quite difficult to do at scale.
On top of those two things, there's been a lot of localization and culture-based work. (For example, photos of people running for fun definitely would not work well in some cultures, accents need to be able to work well, and some words have totally different meanings in different languages.)
Country Code Domains vs. Sub Domains
Right now, Pinterest uses all subdomains (fr.pinterest.com) rather than country-specific domains (pinterest.fr).
A lot of the country-based domains were purchased by squatters, so having consistent top-level domains in every country just wasn't possible, so they use subdomains everywhere.
But if you're a startup that wants to go international, says Casey, you should buy all the country code versions of your name… it's money well spent.
Because Pinterest is stuck using sub domains, for example, their click through rates from search engines is a litter lower than it would be otherwise.
Building a Marketing Team in a Consumer Technology Company
The main thing when hiring a good marketing team, says Casey, is to remember that there are two types of marketing to be done: growth marketing and brand marketing.
Growth marketing is everything surrounding analytics and measurements. It's product-focused initiatives like conversion optimization, SEO, and paid marketing.
Brand marketing, on the other hand, is focused around the company message, positioning, and softer channels like social media and PR, which are less measurable and more story-based.
Typically, these two types of marketing require two very different skill sets. Ideally, you'll want a CMO or marketing lead that understands the value of both. And if you're hiring, you need to know the difference between those skill sets and what value they'll provide.
The earlier you are, the more you'll probably focus on growth marketing, and the more sophisticated your company becomes, the more you'll focus on brand marketing.
Selling the Importance of SEO & Backlinks to Executives Who Don't Get It
Ultimately, you need to prove yourself with data. That's just how most tech companies are run.
Marketing is normally well-understood as a function, says Casey, but technology companies don't have a default respect for the discipline.
With SEO specifically, says Casey, you need to be able to do experiments and show the results.
Backlinks, though, are way more complicated to explain. It's almost impossible to prove the value of link building from an analytical prospective, so you need to do a little theoretical persuasion.
Basically, if you decide not to do link building and your competitors do, you have more of a chance of dying. So if all you need to do is link-build to make sure you don't die against your competitors, it's generally the profitable thing to do.
Growth Tactics Beyond SEO & Email
Casey says that for Pinterest, they also work with publishers and partners.
On Pinterest, every single piece of content comes from somewhere else. So if publishers optimize their content to drive traffic to Pinterest, it ensures that they have a good experience on their site, but also that it can generate clicks back to their content.
Getting publishers to add the Pin It button to their sites, for example, drives brand recognition for Pinterest, but also encourages people to save the content they're reading by creating an account.
But from an over-arching perspective, says Casey, Pinterest grew mostly via word of mouth, especially since it was invitation-only for a long time.
Hiring Good SEOs
Casey admits it can be difficult to find good SEO professionals to hire… particularly in the case of Pinterest, which is operating at a scale that is fairly unique in the SEO challenges they deal with. After all, not many services have over 50 billion pages that are shown to Google.
To help overcome that challenge, he's sought out companies that he sees relatively often when he's searching and recognizes that they're operating internationally and are good at scale. He creates an open dialogue with them as long as they're not competitive to see the things they're seeing.
As far as hiring, he says seeking out people doing really great SEO at other companies has been the most valuable thing to do.
Advice to His 25-Year-Old Self
“Learn to code.”
On a personal level, he'd tell himself to starting thinking about exercise and building a healthy diet. He says 25 is around the age that your body starts telling you things that you can't continue to do at the same rate.
On a professional level:
“Your career is really about the people you meet and the skills you collect.”
He says it's important to focus more on collecting skills, what you're learning today, and how to learn more rather than being preoccupied with a job title or certain role you want.
Weekly Working Structure
On Mondays, Casey does a lot of one-on-one meetings to figure out what people want to accomplish for the week, what challenges they're having, and to uncover any insights he needs to think about. He also sets the tone for the week with a metrics meeting where they look at the previous week, what happened, why it happened, and ways to improve the good and mitigate the bad.
On Tuesdays, he focuses on deep-dives with sub-teams. For example, within growth, they've got an acquisition team, a conversion team, an engagement team, and an international team. Here they break down the metrics on a deeper level and do a deep dive on one specific initiative they're working on.
On Wednesdays, he doesn't do meetings, and the entire growth team focuses on hammering through the large-scale things they want to get accomplished.
On Thursdays, he does a lot of outside chatting (like this podcast), gets perspectives from other people, and meets with the product team to understand other things that are going on.
Four Must-Read Books
- Thinking, Fast and Slow by Daniel Khaneman
- Daniel Khaneman is the founder of a new field called behavioral economics, and he won the nobel price in economics a couple of years ago for his work.
- This book is about the study of how people actually make decisions as opposed to how they're supposed to make decisions according to traditional economics.
- The book breaks it down in a way that's really easy to understand, and shows you that even the most intelligent people make decisions this way, because it's how your brain works naturally.
- The Black Swan by Nassim Taleb
- This book is about understanding that certain things are good at being measured by statistics, and some things are not. It helps you think about problems in a different way, and helps you evaluate both the evidence and lack of evidence that come in after experiments.
- Crossing the Chasm by Geoffrey More
- This books helps you understand how to make progress in growing a business from early adopters to the majority of the target user base.
- It helps you identify where the gaps are, and shows you strategies to overcome them.
- The Goal by Eliyahu Goldratt & Jeff Cox
- This is a book recommended by Jeff Bezos, and it's the best book on conversion optimization that is not about conversion optimization.
- It's one that I (Eric) have forced people to read in the past.
Resources from this interview:
- How to Build a Marketing Team at a Consumer Technology Company
- Why Everyone Link Builds, or Why Sometimes You Do Things When You Don't Know If They Work
- Thinking Fast & Slow by Daniel Khaneman
- The Black Swan by Nassim Taleb
- Crossing the Chasm by Geoffrey More
- The Goal by Eliyahu Goldratt & Jeff Cox
- Casey Winters on Pinterest
- Twitter: @onecaseman
- Casey Accidental
Leave some feedback:
- What should I talk about next? Who should I interview? Please let me know on Twitter or in the comments below.
- Did you enjoy this episode? If so, leave a short review here.
- Subscribe to Growth Everywhere on iTunes.
- Get the non-iTunes RSS feed