Asset 4
Kevan Lee of Buffer Kevan Lee

The Evolution
of Content Strategy

volume icon

About the episode

Buffer has invested in content marketing since its early days, and it’s paying off. But how can content teams keep up with changing expectations and increasing competition for audience attention?

In this episode, we’re speaking with Kevan Lee, Buffer’s VP of marketing. He joined host Matthew Kammerer to talk about the evolution of Buffer’s content strategy and where the team is going next.

You can connect with Kevan on Twitter and his website.

Don’t have 30 minutes to spare? Jump ahead to key sections outlined below, or keep scrolling for a full transcript of the episode.

  • [05:00] Is there a different approach to lead gen vs library content?
  • [06:10] Are there any pieces that are demand gen focused?
  • [09:00] Do you aim to create freemium or paid product users with content?
  • [10:30] How do you compare product qualified leads to marketing qualified leads?
  • [16:00] What metrics do you use to measure content success?
  • [17:35] How do you decide where to spend time with different categories of content?
  • [21:08] How do you align editorial and conversion teams?
  • [22:45] Is there any conflict that comes from the conversion team being reactionary to editorial?
  • [26:21] Did anything change in marketing after Buffer bought back shares from investors?

Episode is brought to you by

  • Matthew Kammerer
    Matthew Kammerer
    Show Host
  • Vanessa King
    Vanessa King
  • Joshua Schnell
    Joshua Schnell

Show Transcript

Matthew: Buffer’s well known for content marketing. Can you share how you started off at Buffer?

Kevan: I joined as the second marketing hire at Buffer. There was one other person who joined a couple of weeks before me, and then our co-founder was running marketing at the time. I joined as a content writer, so I was writing 4 or 5 blog posts a week for our blog about social media.

The two of us, the new hires, were fully content focused. Our whole marketing team was pretty much 100% content marketing back in the day, and we’ve evolved from there.

Matthew: That’s awesome. I was actually just looking at my Buffer profile. I see I joined in August of 2014. It seems like I started paying right around the same time that you started.

Kevan: Oh, that’s great. I’m not sure if I had anything to do with that, but thank you.

Matthew: It was a direct correlation, for sure. So, what does content marketing at Buffer look like today?

Kevan: It has come a long ways, and I’m grateful for the folks on our team who have helped take us there. We do still have the main blog, and content is still our main marketing channel and our main marketing strategy. We have the main blog, which is the social blog, and that includes all of our social media-related content. Then we also have the Open blog, which is about workplace culture, team experiments, kind of an inside look into Buffer. Those have been our primary blogs for a very long time.

I will say there has been lots of pivots along the way. When I first joined we pivoted a week later, so I was writing much different content than I thought I was going to be. Our more recent pivot has been to be more disciplined about who we are writing for on the blog.

So we believe there are two different types of audiences that our content serves. One of them is the early learners or the people looking for one specific problem to solve, and they’ll come to us through Google search. We call this our library content. So looking for a single how-to guide and then they’ll jump off to go do other stuff.

The other audience would be those folks who are looking for more strategies and case studies and news and trending topics and what’s happening now and today. So it might not be evergreen content from our side, but our bet is that it’s content that will bring people back and ideally bring back the types of users who have become quality, long-term customers for us at Buffer. So now we have a library in addition to our social blog.

Matthew: And how do you make the decision on how you split that sort of content?

Kevan: We’ve done a lot of analysis and research into how our content is consumed. What that looks like is going into Google Analytics. Before we made the switch, it was about 80% of our traffic came from search. A lot of that was long-term, evergreen posts that have been around four or five years since I wrote them back in the day. So that stuff has been very long term, very sustainable, high-quality content for us which has been great.

We noticed that the bounce rate is awfully high on that, we noticed that new vs returning visitors had very different percentages. So just seeing this as one particular content type was pretty illuminating for us. We also did some surveys of our subscribers and found the people who subscribed were interested in different content than those who come for one thing and leave. So we realized there really were two different audiences here and two different types of content.

So we started off with some lean experiments. If we cover this particular topic—I think we covered F8 one time—that did pretty well. We covered some breaking news, like Instagram had lots of new features rolling out, so we covered those things when they came out. Did a couple of different case studies, like very specific, tactical blog posts. Those all seemed to resonate with a test group of folks we sought opinions on. So that gave us the confidence to lean a bit further into it.

Matthew: Is there a difference in the way you approach those content types in terms of lead generation? How do you get those library users to stick around and come back?

Kevan: Yeah, we don’t. We’ve come to terms with that. I think our philosophy with that is very trusting. I don’t know how much trust makes sense to include with traditional lead gen and demand gen, but our goal is to build a bias for Buffer. In order to do that, we need to deliver value to folks when they need it, where they need it.

So we’re totally fine with someone coming to our blog, reading something, and leaving us not having any way to bring them back. We have Facebook pixels and other elements we can use, but choose not to. In fact, with GDPR we even took some of those things away recently because we weren’t really using them to any extent.

We’re mostly ok just letting folks leave and trusting that they found the answer they needed at Buffer, and build that bias over time so that when they need a social media solution, we’re the first ones they come to.

Matthew: That makes perfect sense. So outside of the bias building, which I think makes a lot of sense, and I think I’ve experienced that on my own. Are there any pieces that are demand gen focused? Have you given that a test?

Kevan: There was one period in time where we were really lead focused on the blog. We converted all of our CTAs to lead capture forms and were really highly focused on that. It did really well, we doubled our list size in the course of three or four months. We got our list size up over 100,000 pretty quickly, which we were lucky to do with the traffic we get there.

It was interesting. We did that, and we ended up comparing our conversion rate from a visitor to a direct signup against someone who joins us as a lead and makes it all the way through our four or five step lead nurture drip campaigns and finally converts at the final step.

We noticed the drop-off rate from email to email to email built up over time so by the end, it was an order of magnitude less than what the conversion rate was from not even capturing leads and just sticking CTAs on the blog. That was a pretty clear signal to us that we could spend our time better. We don’t need to spend time refining a lead nurture campaign and managing these lists, we can spend time building quality content, getting things to rank on search, getting people to the blog, and that would take care of the conversions itself.

Matthew: From the blog, it’s just pushing right to a pricing page, or is there something in between?

Kevan: We usually push to the homepage. We’ve experimented with how and where to do that. So sometimes we’ll do it from the homepage of the blog, or the main page of the blog with a clear and obvious CTA right at the top.

Recently we’ve shifted to more of just a button that sticks to the header as you scroll up and down. Both of those are our highest converting calls to action on there. And yeah, they link straight to the homepage.

There was a time we linked straight to a business trial, too. So Buffer is a freemium tool, we have a free plan, a pro plan which is $50 a month, and business plans which start at $99 a month. All of them come with a free trial, so we didn’t really see a drop off no matter where we sent folks, whether it was the homepage or the business page. For a time there, we did have people going straight to the business page for our main product.

Matthew: Is the blog generating folks for that freemium tier, the mid tier? Is there a general demographic that you tend to generate from the content or is it a pretty even blend?

Kevan: Yeah, I’d say for the most part it’s an even blend. Acquisition wise, we acquire about 25,000 new users a week. And of those users, maybe 15,000 to 20,000 are free users. They’re definitely the main source of acquisition for us.

But we view the acquisition on marketing as a product qualified lead. So our goal in marketing is to get people using the Buffer product, no matter what tier it is. Then the product team takes that 25,000 number and through a great product experience, in-app updates, and CTAs, they’re able to convert that into revenue at a sustainable rate.

At the moment, we aren’t particular about what kinds of folks we bring in. It does tend to be more on the free side, but even then a lot of businesses start with free plans. We’ve noticed it’s very disparate based on what you’re trying to achieve with Buffer. Some businesses will start on the free plan and use it for awhile just to get a taste of what the dashboard is like and upgrade later on, and others will start on the paid plan. We don’t try to control that too much, we just get people in the door and let the product team kind of take care of the rest.

Matthew: With the product qualified lead, how do you compare that to a marketing or sales qualified lead? Are there different ways to measure that or is that an internal term?

Kevan: I’d say it’s mostly an internal team. We’ve never had a sales team, so we never use sales qualified leads. We are doing a bit more lead qualifying this year, starting in 2019. That’s probably the closest that we’ve come to marketing qualified leads so far.

The way we’re going to do that is there’s a survey we’ll give to folks when they come in, and we want to attract as many of a certain segment of users. Direct to consumer and of a company size between 5 and 500. Previously we have not had that data, so we’re excited to see what effect that has on our acquisition and the knowledge of what content does for acquisition. The best we’ve been able to do with that so far is surveying our blog subscribers, so we haven’t captured that in any other form yet.

Matthew: So are you enriching the leads you get with some sort of external data, or are you asking that on sign up so you’re pre-qualifying by yourself?

Kevan: Yeah, they’ll be self selecting. We aren’t going to ask that until they get to the sign up stage.

Matthew: Cool, that’s interesting. Taking a look at that content, so 80% of it coming from search traffic. There’s another 20% sort of lingering in there. So how do you promote your articles and ensure they reach that target audience?

Kevan: We have two primary ways that we get them in, and that’s search and email. You mentioned search is the big one of course, with 80%. Email we do still have a pretty decent sized list, so we’ll send it out to the email list and get folks there early on.

Those are really the only two things we actively do ourselves or we proactively do ourselves. The rest of it we’re lucky to have built quite a nice flywheel effect so when something new does go out, we have enough people subscribing to the blog through RSS or finding our blog posts on social or through aggregator sites or communities. And a lot of the promotion happens organically in that way, and then we don’t really look at traffic results.

We look at it Day 1, Day 7 and Day 28. We measure success once the 28 days have passed. So we do give ourselves some grace so the SEO can kick in. But yeah, otherwise we don’t do a lot of manual promotion other than sending out an email and making sure it’s optimized for search and social.

Matthew: I want to talk about how you measure it, so remind me to come back to that. But you did say flywheel, and it feels like that’s a relatively new term in terms of marketing funnels. Can you explain it a bit more to listeners?

Kevan: Sure. What flywheel means to us is we have built a set of sustainable actions, and those actions have gotten to a certain scale so that when we publish a new blog post a lot of the momentum is in the system for that blog post to succeed.

In particular for us, that looks like domain authority with search. So we know that if we write a great piece of content and tick all their boxes for SEO, that we stand a pretty good chance of getting our content ranked on page one or two quite quickly. Recently it has been within the course of two or three days, we’re able to see a blog post that we wrote with a keyword in mind on the results there. We use Ahrefs as a tool to check that.

The other flywheel effect is building an email list. As you’re building that flywheel up, if you have a 1,000 person email list at first it’s going to give a certain amount of momentum to that new post but the bigger your list, the more momentum you’ll get. So we have 50,000 to 100,000 depending on how we segment our list, and new posts go out to those folks.

Same with social. We have a decent sized following there that we’ve built up over time. When you combine all these factors, it doesn’t really need much of a nudge for these network effects to take hold and carry the blog posts. Quality content breeds quality leaders and a quality community. It just keeps building over time and eventually there’s less and less we have to do to kickstart that promotion cycle so it feels like a pinwheel. If you blow on it just a little bit, it takes over with momentum. That’s kind of the image that we have.

Matthew: You’re living the dream over there, it didn’t take any hard work to get there, I’m sure.

Kevan: It only took four or five years, yeah.

Matthew: So when you talk about that email, is that a summary? Is it going out at the end of the week with a handful of summary links or is it the full post? Or how are you sending it?

Kevan: We’ve experimented with so many different ways. At the moment, I believe it is a handwritten summary that goes out the day the post is published. We’ve experimented with a weekly roundup of everything that has gone live on the blog over the last week, we’ve experimented with very automated RSS email on Mailchimp, and we’ve experimented with a curated newsletter that includes a link or two of ours along with others.

Honestly, the open rates and click rates of those are pretty equal across the board. We haven’t seen one clear winner. We’re doing it the manual, write a quick recap and send it the day of. More just for branding, we want to be personal and unique and timely and helpful and all of those things. So it hasn’t been metrics driven, it’s been brand driven for us.

Matthew: So when you’re talking about measurement, the two or three check-in points depending on how many days it’s been since publication, what are the metrics for success when it comes to content?

Kevan: We’ve been able to build a benchmark for blog posts, and even that understanding has shifted over time. Initially it was, let’s look at the past six months of content and see the averages for every new piece of content on day 1, day 7, and day 28. I believe day 1 was 7,500 visits, day 7 would be 15,000, and day 28 would be closer to 30,000.

That was great, and those are kind of the metrics that drove us for awhile. Since then we’ve updated it to be kind of a rolling measurement. So, given the last 25 posts that we’ve published, what have been the averages there. We’ve moved that baseline up to be a bit more representative of what’s happened recently.

Another element that we’ve recently started to introduce is understanding that there’s different content types that perform differently. So we basically have two different types. We have the trending, social media driven content that spikes early and tails off late, and there’s not much of a long-tail there. Then you have the SEO-driven content which may start off slow and then gain momentum over time.

Baselining those at the same rates didn’t make as much sense, so we ended up thinking of those in two different buckets and we have different baselines for each of those because they perform differently throughout the day 1, day 7, and day 28 check ins.

Matthew: It’s hard to understand where to invest the most time on those, right? So how do you decide what portion of time is devoted to each of those categories?

Kevan: It has been hard. It’s an interesting spot to be in because we’re lucky that we can decide that and have two different content types. At the moment for us, the decision has been made easier because we already have most of the library content in place. So if you think of the top 25 keywords by volume for social media, we’ve probably written a blog post for 24 of the 25, or maybe all 25 of them already.

The conversation shifts there from what do we need to write to what do we need to update or maintain or keep fresh. That’s a fortunate place to be in. If you’re starting a content strategy from scratch, you’d have to actually write all 25 of those things which would take time. So, because of that we’ve made the decision to focus on the non-evergreen content, which we call the publication content. The trending stories and news and case studies and strategies and tactics. And that’s simply because it’s what we haven’t written about, and what we want to validate a bit more recently.

That’s why we were able to make that decision, but when I started we couldn’t have had both, we could have only had one. We chose the SEO/evergreen one, which I think was the right call.

Matthew: Let’s move in to talking a little bit about team alignment. So I’ve been speaking to a lot of different marketers this season about a recurring theme that seems to keep coming up: the challenge of aligning content marketing and demand gen teams. How is the marketing team at Buffer currently structured?

Kevan: There’s 10 of us, I make 11. We have five different areas: editorial, community, PR, product marketing, and conversion. Conversion is probably the closest we have to demand gen, and then editorial would be the proxy for content marketing.

Editorial is our biggest group, so that includes the two people we have working on our blog, our social media manager, and a lot of people overlap with editorial, too. Our bizdev person overlaps, our partnerships person overlaps, to a certain extent PR and community overlap there. So we have a lot of resources there. We also have an engineer and a designer who help out across the board.

Matthew: Do you get other folks outside of the marketing team to do some writing?

Kevan: Oh yeah, that’s the dream to have an army of writers that you can pull from. We don’t for the social blog, at the moment I’m running our Open Blog, which is our workplace culture and team experiments blog. I am fortunate that people will write for that one. It’s a lot of me noticing something cool that our advocacy or our people team is doing. They’ve been very good to me to humor me with some of those writing requests.

Matthew: How do you get people to write for that? Are you asking them and they’re doing the full outline or you’re saying, I can help you out with this? Or what does the process look like?

Kevan: It depends. I like to offer both ways to folks and let them choose. It typically starts with me identifying what would be a good article, and then reaching out and asking if it’s something they’d be interested in writing and if so, here are the two options: you can write it yourself, or I can get us started and you can jump in. Either way, I need it done by this date. So usually the deadline pressure is good to have them say one way or another. If it’s too much pressure, they’ll let me start with it. I like to leave it up to them, just based on what their comfort level is with content.

Matthew: So, when you’re talking about that conversion team working with the editorial team. How close is that relationship? Are those teams sitting side by side and setting goals in tandem? Is it more reactive based on one side or the other?

Kevan: I’d say it’s pretty reactive, if that were the spectrum. We have shared goals as a marketing team, but ownership of those goals can feel a bit separated out. For instance, in 2018 our goals were on awareness and signups. We had a goal for 150M people reached throughout the year, and 1.5M signups. Ideally those two goals need to connect in some way. Editorial is responsible for the reach number, and the conversion team is responsible for the user number.

What we ended up doing was placing some constraint metrics in between that overlap. So yes we want to reach a huge number of people, but we don’t just want to reach anyone, we want to reach people who will become quality signups for us. Of this reach number, what percentage of reach is coming to the Buffer website? What percentage of people coming to the website are bouncing versus staying? Not only that, but we can track it all the way down through the funnel and say, the the blog sent this many people to our website and this percent converted to paid plans at this rate. So those are deeper, second-level metrics we look at but overall it was the two big goals and ideally that aligned us on working on the same customer journey.

Matthew: Is there any conflict that comes from the conversion team being more reactionary to editorial?

Kevan: That’s a good question. The biggest risk that I’ve noticed is there can be a lack of context to what each team is doing and working on, and they can end up potentially going in different directions and telling different stories.

For example, we have run a lot of paid ads in the past that haven’t been effective for us. I believe one of the reasons for that is the stories we’re telling in the ads don’t match the stories that folks are hearing and seeing when they come to the website itself.

That feels like a disconnect between an editorial team that is trying to increase brand awareness for Buffer and a conversion team that’s trying to optimize a website for a completely different ads audience. Being in those silos can cause some problems with that.

So what we’ve tried to do is simplify our process quite a bit. We don’t run as many paid ads, and if we do we run them much more top of funnel to blog posts and videos rather than to landing pages or the homepage. We’ve been more clear about what this customer journey is. We expect people to come to through search, and a lot of it is branded keyword search. So we’ll rank for terms and pay for terms there that are very brand focused. So what is Buffer, buffer app. And we also understand that people will come to Buffer after knowing a bit about Buffer from our blog, so it’s built this stronger narrative that has allowed us to stay reactionary or separate in that regard. But as time goes on, those two areas will need to come closer and closer together and probably more of a Venn diagram than the two distinct circles as it is today.

Matthew: So you’re fully remote, how do you do that and bring those teams to sit side by side… without sitting side by side?

Kevan: It’s an interesting one. I think we’ve noticed it happen a bit more and more as we expanded from this idea of content marketing to this idea of editorial. What that’s done for us at least, is it’s maybe more semantics than anything, but content marketing a lot of time can give people the perception of a blog. Editorial has expanded that to include content in general, whatever we’re creating basically can fit under that editorial umbrella.

One of the ways that has helped is it allowed editorial to think, okay we want to rank for Facebook scheduling because it’s a high-traffic, high-volume keyword. Does it make sense to rank for that keyword with a blog post or with a landing page on Just by the very nature of that being a question to ask, it has helped bring those teams together because the answer ended up being a page on, and the website is owned by the conversion team so that has to be a conversation there. And the conversion team can think about the effects that has on different funnels and some advanced tracking there.

Just by the nature of that perception changing to be “what is content overall?”, I think that has helped us think about it as one bigger problem or journey to solve for. Practically speaking, there’s tools and things that help us feel like we’re sitting a bit closer together, but ultimately it’s more about those bigger “where are we going”, “what is our strategy” questions that will bring us closer together and not necessarily what we’re working in or how aligned we are on tools or processes or things.

Matthew: I remember reading a lot about the buy back from venture capitalists. I’m curious, did that affect the way you were able to set or think about or execute on strategy from the marketing side?

Kevan: It did not.

Matthew: No major change?

Kevan: Nothing changed, no. Did you read our marketing budget blog?

Matthew: I did, I liked that Carbon Ads mention in there. I wish it was a link.

Kevan: [Laughter] I can change that for you. You’re talking to the right person. To answer your question, we don’t have a very big marketing budget as it is. Buying out investors, yes, it took a lot of our cash flow but it didn’t force us to cut into our future marketing spend in any way. That was probably something that other companies would experience a bit more, so it hasn’t hampered us for 2019 at all with our planning or strategy. If anything it gave us more flexibility and freedom to experiment with new things. We might actually be spending more on marketing than we have in the past.

Matthew: Where do you think content marketing is heading in 2019 and beyond?

Kevan: I’ve been giving that some thought recently. One of the things that we’re looking at is more on the end of experiential content and interactive content. The trends we’re noticing is how popular something like Instagram stories have been where you’re able to thumb through and swipe to consume content at your own pace and style.

We’ve thought about blog posts like “What is code?” from Business Week awhile back, and Typeform’s conversational one. We feel like in order to stand out for content in 2019 and beyond, it’s going to take more than a skyscraper type content of 3,400 or 4,000 words, it’s going to take something pretty special to really capture that attention at scale in a viral way. We’re better on experiential and interactive content being one of those ways.

I think the flip side of that is another trend that we’re noticing in the movement toward micro content or recap content. A website that I’ve followed for a long time has been Pro Football Weekly, which goes out and finds football stories and writes 100 or 200 words on their blog about those stories. If you look at a blog like the Reforge Brief, they go out and source these other great blog posts and give them short recaps and link back to them but what they’re giving their audience is these short nuggets of information that make it much easier to consume content.

If there’s a return to microcontent or microblogging in some way, I think there has been some signs that that’s a more palatable way for people to consume content and build a longer-lasting audience if you’re able to deliver that type of content consistently.

Matthew: What advice do you have for marketers looking to grow in their careers?

Kevan: Always stay curious and follow that curiosity wherever it may lead. I’ve often thought about, should I be designing my career? Should I know exactly where I’m going next? I think there’s lots of value doing that. I’ve been lucky to get to where I am today not having planned or designed anything. I think it’s been purely a curiosity thing for me, where something interests me and I’ll follow that to a certain end.

There can be a lot of value for marketers in their careers if you can diversify yourself. Yes, having a specialty is great but almost having two specialties or many different skills that you can bring to the table, kind of a “T-shaped” marketer is what we call it at Buffer, that’s going to be the most valuable way to grow in your marketing career.

Matthew: Where can listeners find you?

Kevan: You can find me on Twitter @kevanlee and on my website where I write frequently at