Acquisition & SEO Q&A with Matthew Barby from HubSpot

Editor note: For this session, we got a chance to talk to non other than Matthew Barby. Matthew is the Director of Acquisition at HubSpot and is primarly focused on their freemium CRM product. Previously Matt was Head of Global Growth & SEO and is working on something called Traffic Think Tank.

We discussed how to tackle setting acquisition targets when resource constrained as well as from the perspective of a larger growth engine like HubSpot. The video was for the folks who joined, and we're providing a transcript here to anyone who wants to review on their own.


Conrad Wadowski: Welcome everyone, one thing that we were talking about briefly beforehand is how Matthew's responsibilities have shifted over the past year and a half since we last talked.

His new role is now called Director of Acquisition and he is moving into more of a broader responsibility in addition to SEO. So, for the next 20 minutes to 45 minutes, we're going to just chat about how some of those responsibilities have changed, what his recent focuses are, and just kind of learn a little bit from the HubSpot firehose. You have access to so much data and you're deep in this every single day. So super curious to hear that from you.

Matthew Barby: Yeah, I'm excited. And just for everyone listening, when you're thinking about what questions could you possibly ask that I might be able to help with, in all honesty I've touched pretty much every acquisition channel. I've delved into retention and activation as well at some level. It's not just B2B SaaS that I've worked in. I've worked in a bunch of other industries. I have a bunch of side projects that I also operate on.

Any problem where you're like pretty stuck and trying to figure out how to scale up growth in any way, feel free to jump in with some of the broadest questions. If I don't know the answer, I'll be able to point you in the right direction to someone that will.

Conrad Wadowski: Awesome. It's a great way to kick off. So one way to structure te conversation is to focus on a role. One kind of common role is a marketer at a company with say one employee they're workingn with.

I like this scenario, because it helps us start to create a contraint about how to approach the growth problem that forces you to be a little more creative than starting with a huge budget which can be tempting to spend. The majority of folks that are in this kind of situation anyway.

So if you were to take one of these folks right now and say "Hey look, I'm going to create a specific plan for you to execute against". What are the things that you think about when you're developing that plan? And how have maybe that plan changed in the past, let's say, year since we last spoke?

Matthew Barby: Yeah, I think it's important. And I think this actually ties in as well to when thinking about the types of people that we would hire at HubSpot as well, right. It's like looking for certain ways in which you would go about tackling problems is possibly the most important thing for me when I'm trying to scope stuff like that out.

But, when you're thinking about how do I build say a growth plan for this individual's one-person company that we need to get from X to Z, the first thing I think about is: right, what's the dream situation? Let's go agnostic of resources, agnostic of what they can do in terms of budget right now. What is the dream situation of where you want to be in terms of a realistic goal for this company for say a 12 month period and what would you need to realistically do to get to that period. And then we start working backwards.

Conrad Wadowski: So I'll stop you right there. What is a realistic goal? It's a hard thing, right? I'm kind of putting you on the spot here, but if you were to take this type of person, what is something that could either potentially be achieved or is perhaps the right stretch goal?

Matthew Barby: Yeah, and I think what you've got to do here is like every type of business, type of customer they're marketing to, like the persona they're going after, and every industry that they may or may not be in, there is variance here, but let's speak broadly.

The first thing you want to figure out is, with some of the competitors in the space that are both more established and new and upcoming people that are generally. Even if it's tangential to your business. Where do I see most of their general traction coming from?

Are they really strong in the search engines? Quick things I've been using as a proxy for this is website traffic. So I'd use a free tool like SimilarWeb or I'd plug their website into SEMrush and see are they generating quite large proportional numbers in organic search or is it social et cetera?

Let's say we start and we say "Okay most of our traction's probably going to come from organic search". So ranking among the search engines. The next step in there is okay how do I take maybe three or four of the broadest search terms that people are tending to rank for really well and run them through some similar tools, whether it's like Ahrefs, SEMrush, whatever. And figure out what search volume is around those broad topics. And then I'm saying here's potential market opportunity that I could capture.

Now "could capture" is a very loose term because it depends like how well established some of these players are. You're going up against multimillion dollar companies, you may as well assume you have to spend multimillion dollars. This is them when you start to say okay best case scenario, I rank for say a few of these big terms. Break through this traffic and I'm just going to use either some historical conversion rates I have and if I don't even have that and I'm brand new, I'll just use some industry benchmarks. Maybe I'll role off like a typical e-commerce conversion rate of 1 to 1.5 percent, right?

And then I start to figure out what this theoretical funnel starts to look like. Then I can say okay, well if I get to that stage, I'm probably going to be able to think about X amount of revenue, just basing it on an average ASP, right? Average Sales Price.

And then you can start to say okay, well this is where I would feel more comfortable and maybe front loading or taking some of our capital that we may have that we may not want to spend on a speculative bet. Or, more often is the case in this situation, it's less about saying "Hey we're willing to earn less revenue if ... We want to earn X revenue but the area that we're willing to sacrifice is how long it takes us to get there". And that's usually the determining factor that helps decide whether you're going to spend X or Y, right? It's all about time.

And then there's a value and opportunity loss in waiting an extra period of time that factors into this as well. But generally speaking, I like to use just one metric as a general broad proxy for this and search volume is a very good one for that because if people are searching for something a lot in the search engines, there probably is a general demand. There's very few situations where consumer behavior is confined to the search engines. People are often consuming similar information on social media. They're looking for it offline as well, et cetera, et cetera.

Now, let's part that piece and ask another question after you ask the above. Within twelve months, what is the minimum amount of revenue that I'd be happy with this business generating that would enable me to say I'm willing to continue doing this for one more year, right?

What would be too low a number where it would say "Hey, look I'm out"? What then can we do to match up where you need to be in say organic search or generating X amount of traffic from social or be able to get paid working cost per acquisition. For you to hit your minimum "I'm out" stage and then working in that time factor in amongst all of that to suit how quickly you need to be there.

Conrad Wadowski: I like those two ways to back into a goal. In the first case, you're using outside market information. Then in the second you're looking at your internal goals.

And obviously, like you mentioned, it depends on the type of product and context. You mentioned earlier your shift, if I heard correctly to focus on the HubSpot sales product.

The sales product at HubSpot is a freemium product and probably expands into a different type of customer and competitive environment. How did you and are you thinking about that shift?

Matthew Barby: Yeah. So to give everyone a bit of wider context on this, historically like go back maybe two, three years and HubSpot didn't have any real, free products. It was all you come in, we have a strong inside sales operation here, but we would bring in leads via email web forms and they would get passed to the sales team and the sales team would close those leads into straightaway paying customers who are on a free trial setup that you could use and go into primarily just our marketing automation platform, right?

So, I took on, around just under two years ago now, a more of a focus on our CRM product, which we were going after in the sales space, which was a free product. The whole focus was moving our business into a freemium business where we would bring people in as free CRM users and then upsell them into either the paid-for tier of our sales products or over into the marketing tier.

The start of 2017 we decided to go aggressively after freemium and try and transition our business to touchless sales. So just get people to upgrade themselves within the product versus you can bring those free users into a sale track.

This year, we've gone even more aggressive on that and one thing that we had to learn the hard way especially with more of a focus on sales products side of the business is this is a very well established market.

You're going up against the Salesforces, the Zoho, the Microsoft Dynamics of the world that have been here literally since the late '90s. And when you look at Salesforce, literally their stock ticker is $CRM, right, that's what you're going up against. It's an absolute genius SEO player as well.

The way people acquire and make a purchase decision on a sales product is so different to marketing automation software. And the types of people that are acquiring are very, very different. And the other thing is, when there's less of a financial commitment to you, you sign up for a free product, it's very easy to hit those free user numbers but actually be bringing in the whole wrong types of people that aren't activating, aren't monetizing further down the funnel.

So there's a little bit of extra context. And those are some of the challenges. Our typical playbook was to drive most of our growth from organic search primarily through top of the funnel content. We built a growth engine that's millions of visits into our blog that we funnel through into some form of lead capture that we then can get into free products or get into selling them.

With the CRM, that was not working. Sales people that we were going after really didn't have the same kind of time or willingness to consume more educational content like marketeters generally are. And there was also this interesting dynamic where when we were just selling a marketing product we were a marketing automation company selling a marketing product by marketers to markets. It was a perfect story in alignment of trust.

Then all of a sudden we were a marketing automation company selling a sales product to sales people by marketers. Now we're reworking ourselves to be first and foremost a CRM company and we're starting to figure out how to sell that to these different individuals. More importantly what we found is actually we focus on a very small number of really, really important commercial terms.

We rank on page one for the word "CRM" and we went after that aggressively. We rank for the top of page one for "free CRM", we went after that aggressively. Whereas 90 percent of all of our organic search traffic went to the long-tail in the marketing side. 90 percent of all our traffic on the sales side came from five key words, insane ratios. And it's hyper-competitive.

We also needed to diversify where our traffic was coming from at all stage of the funnel. So the biggest area we needed to grow was the middle of the funnel, because with the CRM, more than I would say any other SaaS product, you get a list, the average person will get a list of five, at least, different CRM providers. You will do the most comparison of providers for a CRM product than anything else because it's usually the hub of everything you do as a business.

So we set up and actually I spit out a small team to just focus on acquiring reviews on every kind of review aggregation site like Capterra, G2 Crowd, Software Advice, all of this to go out and influence the marketing team to focus on how to get everyone, every influence in the sales team, in the sales space, talking about us, getting them using the products and actually talking about it. Because we had a solid product and we needed people to trust it when they're comparing us and then focus on building up more comparison content for people that are focusing further down the funnel.

Conrad Wadowski: Right. So from the general HubSpot perspective, you're tasked with building up this entire funnel from the top of the funnel through the end.

Now, I think it'll be helpful to go back to the scenario of a resource contrained executive who's entering into a space that's let's say just an competitive as CRM. Let's say there is the potenital for SEO, but maybe the opportunities without significant resource investments aren't that great, beacuse they've been competed away. The major sources of keywords in this case are already taken and it'd be a hugely resource intensive and risky propositition to go after.

When you're talking about coming in and competing here, what's the thing that's going to differentiate you?

Matthew Barby: Yeah, I think that's where there's an element of one. You've got to be super real with yourself in a space like the CRM space where it's if you think that you're just going to go and build a CRM and hop in and start competing with, even HubSpot let alone Salesforce, you're sorely mistaken unless you have a serious amount of capital to invest. And even then, it takes a ton of time as we've found, right? We're a public company, we have capital, but it still took us a long time to compete.

That said, where I see people able to break into this space was less been like "Hey, we're CRM" ... The other thing is for all intents and purposes the CRM space is very difficult to truly innovate what a CRM is. There's some very clear things that people want from a CRM and it's tough to innovate around that. So the key area of opportunity is actually more on a land and expand model of going after a specific point solution.

There are a few companies I think have done this particularly well and I'll give a couple different examples of lesser known companies that have attacked in the sales space and are working their way outwards.

One example of this could be Drift which came from former HubSpotters. They have a live chat product, another key area of the sales space, an area that we're attacking now as well.

But instead of them saying "Hey, we're going to build a live chat product into a big CRM platform", it's like "No. We're just going to attack the live chat space and we're going to get really, really good at that". The thing that I think about a lot and the biggest challenge that we face on acquisition point of view is not how do we beat Salesforce.

I actually feel pretty good about that, like longterm, that we can win out there. The biggest challenge is how do we beat people who just create straight up point solutions that go head-to-head with one feature. At the end of the day, if your company's whole focus is around one feature point solution it should always be better than a single feature of HubSpot's platform. Otherwise you don't deserve to compete in the space.

Then you have things Calendly or Mixmax if you use meeting scheduling apps, right? And they're just like "We're going to be a meeting scheduling app and we're going to be killer". We have a meeting scheduling app. Are we as good as Calendly and Mixmax? We're getting there. But we have to go head-to-head with resources with we probably spend one 30th of our time focusing on how do we attack that space and they spend 100 percent of their time.

I'd say Calendly have done a particularly good job of this. And they capture all of this demand in an area we're only semi-focused on. And then they can build out, right? They establish themselves in that space, create this moat around one feature set, and then they have to do the difficult job though of how do we then do this again for not a point solution. They expand that way, but they've built this sustainable revenue model that's coming and brand recognition around this one individual feature.

It's a lot easier to gain penetration in a market through a point solution acquisition route, but long term it's much more difficult to expand into a platform. It's much more difficult to start and build a platform acquisition model, but it's much easier then to attack individual point solutions once you've got yourself there because you own the entire platform space.

Conrad Wadowski: Just to pile on, another consideration along with this point solution wedge, to get even more narrow is competing in a specific channel. You can say "Hey look, we're not only going to be the top point solution on calendaring, we're the top calendaring app on Facebook".

And maybe there's not a whole lot users who are thinking about calendar solutions on Facebook, but the biggest Facebook group on calendaring solutions gets you to the next step.

Matthew Barby: That's a really good point that I didn't touch on as well it's because we also have these added layers as well where people typically point solutions do very good jobs of carving out unique channel opportunities as well.

Virality is an important play in a lot of that space. I'll give you an example with Typeform for example, the form building company. They have a viral loop baked into that product where if I sign up for a Typeform account, I'm likely to then create a form and it has on the form at the bottom "Not using Typeform? Use it here". And one of my survey responders will see that and potentially create their own Typeform account. Similar with Calendly, they have "Not using Calendly yet? Here's go create your own thing and schedule your own meetings now".

We have someone on my team who's just dedicated to just solving viral loops within our product. But it's an area especially within our sales space where people have carved out whole businesses. Calendly and Typeform it's probably their number one acquisition driver.

Conrad Wadowski: Right. And maybe to take a step back, we'll talk about actually just how to manage these kind of experiments maybe a little bit.

But before that, let's talk about some of these newer channels, or maybe not newer channels, but alternative channels, to really narrow as a way to create some sort of a wedge in some part of the market.

The first thing that actually one of the GrowHack members asked was voice and the newish behavior of asking questions on Google. When you think about let's say voice, is that a channel that you've thought much about or are experimenting with?

Matthew Barby: For sure. Yes and yes. What I also try and do within the team is we invest a lot of our time into big bets. So every year we specifically put within some of our goals along the more quantitative stuff, like we have to take two to three big bets one being a brand new channel.

Last year we opened up an affiliate program, which was a big bet that we're working on bringing into fruition now. Another big bet was an early MVP for acquisition through chat bots, which was a super fun project to work on. And we are really starting to make that work.

The chat bot acquisition play we made led us more into exploring how do we understand how different platforms and different interfaces to learn how a user will interact. And a big focus actually at the start of this year is how do we prepare ourselves for voice. And it's quite tough. In all honesty, I've seen people talk about this and people giving like grand strategies right now for appearing in voice and it's largely nonsense because that space is completely still developing.

But one of the biggest areas we've invested from an SEO point of view is figuring out featured snippets. That quick answer box within Google search, we've had some very good success with some experiments we ran last year. I think a lot of the stuff that we've found from that is going inform a lot of the things that are going to happen within voice.

There are a couple of things that we know, right? One, the search engine results page is going to become one result instead of 10 on page one when you're talking about voice because Google decides, hey this is the number one listing and I'm going to read it out to you.

The next thing that's going to happen is how do I structure data in a way that these voice platforms can actually go through and consume my information. I actually start to see some messages coming through Google search console recently. I got one myself that was like "Hey, markup some of your content if you have a podcast site, a recipe site, or a news site to focus on specific tags for Google Assistant". I think that's going to adapt more and there's a whole actions library I believe they call it now in Google Assistant that's going to be a whole new area of markup that's going to be rich snippet markup.

We've been playing around a lot with Alexa and building some sounder Alexa skills just to figure out that whole marketplace. I personally think, and this is one of the bets that I've been kind of making over the past year and will continue to share is that what we're going to start to see is the fragmentation of search platforms. I think people more and more ... I don't know if you, Conrad, if you use Alexa or Google Assistant. I use Alexa.

Conrad Wadowski: Personally, I use both of them. I'm kind of testing assistant on my iPhone, and then Alexa at home.

Matthew Barby: Right. And I think one of the things that's apparently kind of like the dawn of mobile apps, we all moved a lot of our user behavior into an app store. I think that's one big thing that we're starting to see with Alexa and Google Assistant is ultimately these are self-serving walled gardens that people can start operating within. And I think one aspect of this is less just about how do I optimize my web content for voice and is like how do I be within these ecosystems and be a trusted entity within them. And I don't have all the answers to that right now, but what we're just trying to do is let's just make bets on, do we need to build our Alexa skills, do we need to markup our content with new speculative markup, do we need to refine all of our different place where we can be seen as an entity in Google. We've been going through a big thing of updating a lot of Wikipedia pages, updating Wikidata. All of this stuff to try and make it as easy possible for voice and the various platforms and search engines to consume and deliver our content in a format that they choose.

Conrad Wadowski: Ok cool, so when you're doing these kind of experiments how do you think about running them? Are you looking for let's say an outside expert to, I don't know, get you up to speed on a particular acquisition channel? Do you have someone on the team that goes out and does the basic research? What's the way that you're structuring these bets internally?

Matthew Barby: There's a few different ways that we go about this. I would probably bucket these into maybe three different distinct groups.

One is someone has an idea in the team and they're willing to back themselves and say "Hey, I've done a little bit of research here and I need X amount of resources to dig into this initially" and they say "Okay, this is what I need to find out this date for me to viably continue looking into this". If not, we kill it. And then someone else may explore it in the future, but we cannot dedicate enough time to that. That's the most common used case of new things.

The second piece is we split out a full team. And this is kind of like a second step. Sometimes for this, we did this with chat bots. We run some small tests. I was involved in this, James Gilbert who runs our APAC office was involved with this and we just spun up some chat bots, like random stuff right at the start of last year and just started to see: one, how to people engage with chat bots and how does it differ across different platforms. Our Co-Founder Darmesh created our GrowthBot which we started pulling data in as well.

And then we were like, "Could we actually use chat bots to do lead qualification and maybe even generate leads?" We opened up a Japan office in the past two years and funny pretty much all the rules in Japan are the opposite. That's a whole other story, but they use Facebook Messenger for business a lot. So we really started shifting some of our focus in Japan on how do we drive acquisition with chat bots.

Then the third bucket is "Okay this is interesting, we need way more resources", split out a team, set new goals, and now we actually have a team of four full-time people who are dedicated.And then the third piece would maybe be, "Okay, we have this kind of broad idea. Yeah, maybe we need to chat to a few people who are experts in this or maybe we need to just straight up go in and acquire some kind of entity", whether it's someone's business, someone's website, someone's app, or whatever and dig deeper into this.

And then ultimately we'll go back to, "Okay, then after we get this knowledge, what's the next steps?" and then it goes back to step one and then step two again of what I was just saying.

Conrad Wadowski: Great. And that makes a lot of sense in that a lot of companies have challenges with running experiments on these newer channels because it's risky and most of them are probably going to fail.

To someone that says to you "Hey look, this is probably not going to work", what do you say to the person that is internally at a company, let's say, and maybe it's not like this at HubSpot, maybe with some of the client work that you've done, to someone that says "Hey look, the chances of this experiment working are, I don't know, less than 50 percent, maybe significantly less", how do you justify that to an internal stakeholder who wants to make sure everything is working great?

Matthew Barby: For sure. So there's a couple of things there. One of the things that I set within our team as a benchmark is that 90 percent of everything we test will fail. And the most important thing is at the end of the day, if you're in acquisition marketing of all things, you have to be used to failure and you have to deal with failure in the right way.

Every single test that you do should be never be determined. For example, if you are trying to dramatically grow revenue by exploring a new channel, your first thing to validate whether something is worth even exploring at the second version should not be "Hey, we want to grow revenue from this small experiment by X" it should just be like "Hey, let's start up with something very small with very small outcomes". What I'll often say is get a glimmer of hope.

And it can be a very unscientific thing you do. Low resource, low input, and low outcome potential. And then that's when you can start saying "Okay, now we're going to create a baseline benchmark to scale these things up". But you should no matter what always come out of something where you can be able to take that learning and either apply it to something else, or just get yourself back on the right track.

At HubSpot, one of the reasons why we do so many experiments is not because we want to test and find new things. Actually, I actively discourage that because it can sidetrack you a lot, trying to find the new things when actually what you just need to do is things that you're doing relatively good at right now better, right? One of our focuses is: we're very aware that we're pretty heavily dependent on organic search, right?

It's a nice problem because we've just done so very well, but we wanted to try and dilute risk out, right? Let's imagine everything moves to voice tomorrow. This is obviously not going to happen, but that would put us at significant risk. So we actively run campaigns to mitigate risk in this sense and proactively solve problems in our funnel that might come up.

In that situation, experimentation is important. For a small business, I would say less than five percent of your time should be spent exploring new things because what you need to spend time on is "Hey, I just need to figure out what's that one channel that I can just dedicate my time to that I can scale up some growth" and then go super heavy on just figuring it out.

The worst thing that I see a lot of small businesses fall into the trap of is trying to solve their entire acquisition funnel and build up this huge growth engine within the first twelve months because actually that seems like a long time to a lot of people that are wearing many hats in the early stage of the company. But actually that is such a small stage and will end up being very insignificant in the whole lifetime of the company when you're like 10 years in and you've been successful, which is probably your endgame, whether you're looking to make an exit or whether you're looking to grow a sustainable business, right?

Conrad Wadowski: Makes a lot of sense. If you're smaller you have to balance this risk of experimenting in a newer channel with lower risk bets. That might mean amplifying what's working for you already or be going into a channel you can learn a lot about.

So, the last question I'll ask, and I'll move on to some other questions, is going to be about video. One thing that we've started doing a lot of, and not yet for acquisition reasons, at Ops is generating a lot of video content. I know HubSpot has run some experiments using video for acqusition and SEO. How are you thinking about video, and is that something that HubSpot investing in?

Matthew Barby: Yeah. We've had a very love-hate relationship with video at HubSpot. I think one of the things that I've learnt myself over the years with video more than any content type or format, should I say, is that ultimately if your users don't enjoy video, you will never convince them that they could enjoy video. It's one of those things where it's very easy to make a case internally when you're building up content that a video would be really great to explain this thing. But that's largely because of a failure on your side of being able to actually explain the thing you're trying to explain in a format that your audience actually wants to have content delivered in.

And this is a much broader spectrum and I think this is like still having remnants of when infographics took off and everyone just wanted to build an infographic and it's like "Hey, by the way people don't want to consume this information in this way". You need to go for what they want to consume in.

That's the first starting point of video. And that's actually where historically HubSpot, we went wrong there. The second piece that I found that's incredibly important is do not ignore the importance of video and video platforms like YouTube. One of the things we did was "Hey look, we want to like build video at the top of the funnel to drive traffic" and we said, let's create some nice custom stuff on Wistia and not going to push anything on YouTube. Maybe we'll gate some of these videos because they're super high value.

In all honesty, if you're doing video you have to be on YouTube if you're interested in the top of the funnel. For some of these pieces of content or Vimeo if you're in that right space or some other niche video platform. The discovery engines, kind of like Medium, right? Like For certain industries, articles rarely rank in search engines, but all their traffic comes from this internal community and this referral engine that they've built. Same with YouTube.

There's a very specific type of content that works really well on YouTube and unless you are using video to basically engage an existing audience, I think there's very few times you should be like really dedicating a bunch of your time to solving like really appearing really well in YouTube.That's one of the two big pieces for me in video is making sure you actually figure this stuff out from an acquisition point of view and you need to be able to very quickly pivot away if things aren't working out correctly as well.

Conrad Wadowski: Absolutely, so let's go to a couple of different questions here on competitive collateral. And I think that's a really interesting question because I think that there are a lot of different products that are out there. And often times the category these days is pretty well defined. You're looking for a calendar app. So you already know that the demand is already there for a calendar app. You might already be using a calendar app. But it takes a lot of time to look at let's say the top five or top 10 or top 20 different calendar apps.

What's the way that you think about that kind of competitive collateral? And have you focused at all on that in the recent period of time?

Matthew Barby: Yeah, it's funny. It's so fresh in my mind right now because we just brought someone else onto the team and a portion of their time is completely dedicated to figuring this piece out. In SaaS, competitive collateral and owning competitive search is just so, so important. It's quite difficult on some levels to measure the impact as well because a lot of the time people don't necessarily convert on competitive collateral, but whoa do they 100 percent check it during their decision making process. It's the tab that's open always when they're comparing vendors and often discovering vendors. And there's a few layers to this, generally speaking.

So, if I think about ... I'll use general SaaS products, right, if we just stick to say CRM because it's super broad it can be applied to pretty much anything.

First of all, you've got the general software name, and then the word "software", right? It's like "CRM Software". That's when people will largely come through onto a vendor aggregator site. That's your G2 Crowd's, your Capterra's.

Conrad Wadowski: Right, there are a bunch of these ad-based business models.

Matthew Barby: They're an enormous acquisition driver for anyone in the SaaS space. But it's just important more than anything. Let's just take actually traffic that it drives which will still be a sizable amount. You need to be there and you need to have a strategy there for either paying to play there or in some of the ones that aren't pay-to-play. I think G2Crowd is one of the few that aren't actually pay-to-play and it's more about gathering reviews in there. But having a review acquisition strategy is a massive one.

One thing that we did with a lot of these companies is bring them into our inbound conference that we run where we have tons of HubSpot customers there and get their team focused on speaking to our customers to actually go through and leave reviews there. And then we've also done stuff where we just have email launch mission go out and so people continuously will leave reviews. But that to one side, that's like the top piece. It's like the broad "I'm going to find a vendor" term.

One layer down is the "best" software name. It's like, "best CRM software". You can actually compete usually with Capterra in that space and we always have. If you look at our CRM product pages, we always have a /best page, and what that page will do is talk about how awesome HubSpot is, right, but then it will lead you into you'll have a grid where it's like a competitor comparison grid. It will just be basic feature stacks, but then link into the next layer down which is like say HubSpot versus Zoho.

Now, the important thing that we do here is we always do, as opposed to saying HubSpot versus Zoho, we'll do Zoho versus HubSpot, and that's more the taxonomy we want to rank for because what we want to do is look at people who are currently evaluating Zoho against other vendors. So we'll capture demand from other people's decision making funnel and we go after that and you can scale that pretty well and then just have like a bunch of different stuff in there. And that's where you get down to the vendor versus vendor. You can get even more advanced with that with vendor versus vendor versus vendor. And you can start to imagine how you scale that content format with a simple template.

And then the final piece of all of this is your individual HubSpot CRM review. You need to have a page on your site that you own where you can collate stuff and wherever possible try and own that ranking. It's usually really difficult and also doing any reputation management if you need to in amongst all of that, that for me is like your playbook in amongst all of this.

Conrad Wadowski: Are you using an outside vendor right now to handle some of those reviews?

Matthew Barby: No. We manage that all internally. We have our partner marketing team, we'll have our customer success team and we also then last year actually we focused on review acquisition directly within the SEO team. A little process that's now like a growth engine that we've got. We've got email automation going out, working with the different platforms, and we'll do some incentives to just get people to actually take an action and stuff like that.

Conrad Wadowski: No, I think that's some really good takeaways there just in terms of almost generating a lot of content from your current customers is one way to differentiate with some of the competitors and you could use that inside of any kind of SEO or any kind of acquisition you're doing against other competitors.

Matthew Barby: 100 percent.

Conrad Wadowski: We got a couple more questions here. What would be your tips of helping a larger content site which is already getting tons of traffic from search, so 2.5 million, but has reached SEO plateau? So this is kind of maybe in a similar situation to what you were at at HubSpot.

Matthew Barby: Yeah, I mean, I've worked with companies that are big publishers that are more like blogs attached to sites and similar kind of sphere. I tend to bracket it into focusing on two broad things.

One, where is all of the traffic coming from right now, and it seems like the large amount here is organic search. So then we'll go one layer deeper. It's like "Okay, what's the variance or the deviation in the amount of different queries that your ... So basically what I'm saying here is does 90 percent of all your traffic come from say three percent of search queries or do you have a really wide spread of those? And of the queries that drive the most combined traffic, are there any patterns that you tend to find in this?

One big thing that we've found at the start of last year at HubSpot was we're seeing a little bit of a plateau across some of our blog which we've never seen before and we drilled it down and down and down and was like "It's a lot of these how, why, what, where type questions. Okay, we're starting to see a pattern there. Our are rankings dropping?" And we realized, yeah we were still ranking number one, but there was these featured snippets were then appearing and we weren't ranking in those featured snippets. And it was causing a huge issue for us where it still looked like we were ranking number one for a bunch of these keywords but actually there was like this position zero some people are calling it now, right, ranking above us more importantly taking huge amounts of the click through rate.

For anyone that's running a big content site right now and has seen a plateau over the past six months, I usually say that's your number one thing to look into. And what we now have is we have someone dedicated on our content team to featured snippet optimization where they'll go in and add short snippets of text. We've actually wrote about this on the HubSpot blog which you can check out on how we do it. And we've dramatically grown the amount of featured snippets we've grown.

I take that similar playbook and write: okay, if it's not featured snippets, is it two or three big posts that have seen drops and that's why? Or is this a more of a fundamental content is not resonating with our audience? Or, which is another piece which we started to notice, did we over the past three to six years jump on some interesting emerging trends that are no longer emerging, right? HubSpot inbound marketing. Look at Google Trends, right? And email marketing and social media marketing. They're all like we jumped on the very beginning. We got in here and now it's plateaued.

Now we need to find what are the new emerging trends. So we're thinking about how do we create content about VR, AR, blockchain, chat bots, all these things that could potentially drive huge growth for us over the next three to four years and we need to start investing in that now.

Conrad Wadowski: When you're thinking about going into one of those spaces, we were just talking about, let's say VR or something like that, what's the process that you're evaluating a writer or getting someone that's an expert. Are you going out and interviewing experts? Or what's the way that you're acquiring that information?

Matthew Barby: Yeah. I think there's three routes that we could go down. One, we work with a freelancer whose maybe running a blog in that space that we think is quite cool. To get this up and running and off the ground before making a bigger commitment.

Two, we take that said influencer and we just acquire their blog and them and then absorb that in. That's like a nice advantage that we have because we have capital to do that and that's what we should be doing is using that as a force for driving acquisition.

The third thing is "Hey, do we hire someone in who is fantastic researcher and can learn a lot of this stuff?" And we maybe do some consultative stuff while we pair up a freelancer with that person, get them up to speed and then they could just roll with it. That depends on the topic. The other thing that we often have as well as like, okay, VR, AR, we have Austin Knight are head of UX at HubSpot is a thought leader in VR and AR. So we can sit Austin with these people and learn from them. It's like blockchain ... I swear I got a bunch of people, myself included, who have been heavily involved in that space and can go and help bring people and skill them up. When we started going into working in advanced SEO side of things where people like Victor, our head of SEO, can go really deep and help the content team take really tough concepts to understand and distill them down and go in.

But, I think this is all about "Okay, what's the minimum viable thing we can do to prove this out? And then how do we scale up?"

Conrad Wadowski: Got it. This is an interesting question. What do you think about B2B cold email as a way to prospect for small digital agency?

I'm just curious on your take on this because I've gotten so many of these emails, right, and I don't personally respond to them. I don't think that the email outreach is well researched enough or relevant enough often times. But yeah, what's your experience. I mean, you're getting these emails right now. How are you handling this?

Matthew Barby: For sure. I can already see the tweets coming in now. Like HubSpot's Director of Acquisition promotes cold-calling and cold-emailing. No, I think this is one of the things, and this is actually one of the things that we learned pretty strongly when we started marketing to sales people was we really wanted to go in with a mantra of "Okay, like this idea of inbound sales and the cold-calling is the devil and cold-emailing is the devil". The reality that we found is whilst we don't want to be progressing these areas and we want to find solutions, they do work.

They do work very effectively if done correctly. The thing for me is like is this sustainable? Is this a channel where when you perfect it, it dramatically drives down your CAC, your customer acquisition costs with it, not really, right. It has, from the moment you begin it, diminishing returns. So, that's a channel where I say okay ... If we're talking never start, and this is something that you can scale growth, it's like "Alright, okay, there's so many case studies where this has worked". You can argue that Airbnb's Craiglist plague was a similar thing, right, and that's what gave them early traction. Would it work right now? No, right?

And that's the other thing that factors into a lot of this stuff and it's a huge thing with SEO is tactics ... Every tactic expires, every single one. From the moment it was effective it begins to diminishing in value, right? You've got other people abusing it, the effectiveness of you being able to scale and reach net new people and then also the general industry just ruining it for everyone.

So there's that kind of play and I would just factor that in. I would never say that anything doesn't work because it does in some circumstances.

Conrad Wadowski: Yeah, absolutely. Kind of a quick aside. Our head of growth at Teachable came to us through a cold-email and his email actually linked to a blog post that was, I think it was probably around 10 pages maybe longer, about what we should do to improve. And now, he didn't have all the information, but it was a great way to get a sense of his thinking process.

Matthew Barby: I actually had a similar example. I just started a whole new project and it's going to be a big content project that we've been doing. And I was looking for an editor, like a proofer and editor for all my work and this guy randomly just cold-emails me with this two page critique or a recent article that I put together with the finest details with all this stuff, and he was like "Have this for free. Take a look. Like, I'd love to work with you in some capacity". And it was complete cold reach out but it was so high quality that I've spent a bunch with him now and like a core part of my team and he's awesome. So yeah, it can definitely work.

Conrad Wadowski: Awesome. So we got a minute left, so I just wanted to say thank you to Matt. If someone wants to followup with you, what would be the best way to do that?

Matthew Barby: Yeah, you can reach out to me either through my website,, new project that we're launching that's a big online community to grow and learn. You've got Twitter as well @matthewbarby.

Conrad Wadowski: Awesome, thanks again Matt and we will have a recording of this and we'll send it out afterwards.

Matthew Barby: Yeah. It was great Conrad. Good catching up. Thanks everyone for joining.

Matthew Barby

Matthew Barby
Director of Acquisition at HubSpot