Evolving your Approach to SaaS Marketing
Jim Szafranski, COO, Prezi
Jim brings a decade and a half of experience of successful SaaS marketing. He had the opportunity to lead Fiberlink’s product, marketing, and customer success platforms as well as help create an innovative, high-velocity sales operation that delivered nearly 20% monthly growth for 4 years prior to IBM acquiring Fiberlink in 2013. As an executive IBM, Jim led the scaling of Fiberlink’s SaaS customer acquisition platform globally and expanded mobile industry partnerships with Apple, Google, and Microsoft. Now at Prezi as COO, Jim’s role includes leading Marketing, Sales, Success, and Analytics.
View transcript
Hi, thanks. Good morning. Thanks for having me. You can see the topic. So the way I was going to start was transport yourself back to 1999. And imagine you're searching for an online sales contact management tool. You go to this site called Yahoo and you browse the utility category and you find this. This is salesforces.com's 1999 page. And I'll use this as sort of the beginning of our talk about evolving our SaaS marketing strategies and talk a little bit about the evolution I've gone on and what we've been doing at Prezi and how we're trying to stay up with things. If you think back, marketing was pretty easy back then. your information online and at low cost. That was kind of it for SaaS marketing in the beginning. For me, this played out probably for about five years. You could just get to the top of the hill and shout, we're online and we're low cost, and people came. And as we were doing that, I think most of us got wiser that SaaS gave us an engagement model that was different. And so we were able to focus on getting organic search rankings and getting clicks and trying to drive people to product trials. And as we did that, we're like, well, this is working really well, so let's do more. So we started doing things like paid search and eventually digital stuff around social and things like that. That got expensive. And so we were like, how can we get in control of our cost and our clicks? So we started doing content and email marketing. I know there's some good talks around those types of topics here. And we got into this digital era where beyond just low costs and your information online, we started talking about time to value. We can speed you up. You can come online, you can learn something, you can click on it, we'll get you going. SaaS is great. And as we did this, what I think many of us have seen is the marketer, the CMO, started to be really tied to revenue. In Silicon Valley today, many CMOs think of themselves as the driver of revenue. And so we started inventing things. In some cases, we started running inside sales and having sales development reps. And we're working with all these clicks. What do we do with these things? So we're running salespeople. And that tie back, the way that all this stuff tied, led to how are we going to measure ourselves? In the beginning, it was the magic number. And nowadays, it's customer acquisition costs. It's gotten very analytical. I can get my clicks. I can get up and try things. I can have inside salespeople. What are our costs? And the cost of acquisition, what we've seen is most people think that, hey, if I spend the same amount of money to acquire a customer as they spend on me in one year, that's pretty good. That's SaaS. It's all about the future. But then the future started mattering. So we started worrying about retention marketing. How do I engage my users? We introduced different metrics like long-term value or lifetime value of a customer so that we could tailor our marketing on a daily basis to which users and which channels are going to stick around longer. Because if my cost to acquire is equal to my first year bookings, I need these people to stick around. And so we've done all this. And this has been the journey that I personally have been on and I'm sure many of you have been on. And so where does this leave us today? And I think as opposed to in 1999, hey, it's online, it's low cost, it's pretty darn complex now. We have all these techniques and all these tools stacked on top of each other. And for me, the one way I envision this complexity is this programmatic advertising. To me, primarily it kind of works. But we don't know why. We don't know how it works. And we don't actually know how long it's going to work. It works for now until we're all doing it. Then it's not going to work again. And so the other thing that's emerged for me is not just complexity but interdependence. There's a high level of interdependence for me in the marketing function and for us at Prezi across all kinds of functions. And marketing is always, business has always worked together. To bring an example here, we're talking about a lot of social advertising and things like that. Well, if you're going to advertise on social, your product probably needs to log in on social then. Why would you advertise on social and then have people have to create a new account? That's a waste. You're going to lose people. Well, then if you're going to let people log in on social, then your whole support team needs to be supporting users on social. Because if they're logging in on social and you're advertising on social, they're going to bitch on social. So you have to get back to them. But then if you're doing that, you can see how about if it's an enterprise sale? Well, the enterprise, the person who learned about you on social, they're going to go look at the tweets and the LinkedIn page of all your salespeople. So then the salespeople have to have the right content on their social feeds. It's so interdependent. Who thinks I'm going to spend 25 cents, I'm going to advertise on LinkedIn or Twitter? That's what we're thinking. But if you look behind the scenes, you need all these things integrated. Or don't spend the money. The funnels are just going to pinch and you're going to lose these users. And so there's no silver bullet in my opinion, in our opinion. We have to do all this as marketers today. And so what I'll try to do is spend about 10 minutes talking about how we're doing it at Prezi and at least how we're thinking about some of the futures too. And then if there's some Q&A, we can get into that. And the way I'll do that is I'll highlight six key things. There's a lot we do. Before I do that, though, again, I gave you that example. This history, we talk a lot about integrated marketing. We need all of our channels, our content, everything to be integrated with the messaging. And I like to think of it as it's integration marketing if you think about that social example. Again, why would you send a tweet advertising your product if you're not social across all these functions? And so some of the things that we're focusing on at Prezi today in marketing. So first, you have to know your data because all this stuff is so interconnected. And the way we think of our data, we think of our traffic, we think of our conversions, we think of our behaviors inside the product of those people we're acquiring, and then also what messages and what problems, what value statements are working. And we track this daily. And why do we do it? Because we want to know it's working. We want to repeat the stuff that's working. And if we're lucky enough to figure out why it's working, we try to replicate it in other channels. And so on a daily basis, we know every visitor to our website. That's probably about four or five million a day. Where are they coming from? Which channels are they coming from? Is it mobile? Is it iOS? Is it Android? Is it LATAM? Is it Europe? Where is it coming from? And we're trying to track these trends. And not to go too far, but this blue line spiking up. That's our Spanish speaking, primarily South American traffic. Holy cow, it's going up. I've got like 50 million more people visiting us in that region. So what do we do? We have to find people in region. As opposed to just making the language work. We have to basically start thinking about how do we market and how do we sell in region for those things. So we're making changes daily and things like that. Conversions we're constantly tracking every campaign. Blue lines control, orange line is the campaign. But we're constantly tracking what campaigns are working. And then whenever we find ones that are, spikes, we double click into that. And then we try to repeat. And we're reviewing this stuff with our board every month. Which campaigns are working and why? The nice thing is that they share back. Well here's what Prezi is doing, here's what Slack is doing, here's what's working. Let's keep going. We don't stop at acquiring the customers. We think about the product. Okay, so we acquired this customer through this tweet. We're trying to do onboarding and training now via Twitter. We're actually paying, let's say. And we look for these disruptions in our products. So marketing doesn't stop at marketing and CAC. Then it's what are the users doing and should we get more of the users or not and how are we finding these splits of behavior. This represents about like 5% more engagement in our product based on one thing we tried. And it works. So then we keep kind of doubling down on that blue line. And marketing is figuring this out. Marketing is saying I paid to find these users, let's see how they're working in our product. And then we're constantly tracking what messages are working. And so these are the different messages in one of our campaigns. Surprise cloud works. But we're constantly trying to figure out which messages are working for which cloud. And then the size of these bubbles, if you're curious, is how much available inventory is there for these messages. So if cloud's working, also can we find more inventory wherever we were making that work. So we're constantly not just which messages are working but how much inventory is out there for us. And so that's how we kind of think about knowing our data. Again, we have analysts embedded in a team. We've got two analysts embedded in the team. We log everything. We try to do attribution to everything. And we're not great. I mean, we make mistakes. Sometimes you're like, is that data really right? But we really start with the data. Another thing that's been key for us as we continue to drive into programmatic ad buying and all these things is this old thing called SEO. It's an amazing cost for marketing. We always want more budgets. And so last year, at the beginning of last year, we were doing a lot more digital, more search engine marketing, more paid search. This is our Alexa ranking. And the opposite, lower is better. And our organic traffic started going down because we were diverting money and spending money on paid. So we said, hey, that's not right. So we pushed on SEO. I don't know when the last time we had done it at Prezi, but we just started cleaning up our site maps. And we just started really doubling down on our SEO. And you can see where our website from an Alexa ranking perspective started really improved. And then we made a huge push on mobile at the end of last year. And so we've gone from the 1,500 visited website in the world to like 640. This cost us like, I don't know, half a million dollars. If you look at it on what would have paid to do this, it's something like $15, $20 million. It would have cost us to get this amount of traffic. And all we had to do was improve our site maps and make some of our marketing and our product work a little better on mobile. It's insane. It's a crack for our P&L. It's free. And what we've done, this chart's no good, but we've gone from our organic search going from 58% of our traffic to 72%. So again, you can see where that would just have a huge boost on your P&L. You double, triple your marketing budget without doing much. So we've got the data. We're doing the basics on SEO. Content ends up being, again, continues to be a key thing. We were talking about last night, even like just stupid webinars, right? Simple webinars. Everyone's doing webinars. But it's amazing things like that work, especially in the way we think of it. Before I get there, why does content work? Has anybody read this book? It's called To Sell is Human by Daniel Pink. So what he talks about is we're no longer in the era of buyer beware. Have you guys heard that term, buyer beware? Slick salespeople, they hold all the information. They know all the negotiating strategy. They trick you to buy a bad car. That doesn't work anymore, right? He talks about the era of seller beware. Your customers know more than you about your product. They know more about their problems than you do. The customer knows everything. And so if we think of that as the mindset, then you have to... Customers are constantly looking for education on how to solve their problems. And so as a marketer, the more you educate them, it seems to work. That's been my experience. Here's a funny juxtaposition maybe for next year. One of the things that's happening in this era, we're getting all these clicks. We're hiring all these entry-level junior people. We're giving them a script and we're saying follow up with these people. We're hiring everyone. They know less than the customers they're calling. And it's a weird juxtaposition, this whole inside sales and sales development rep thing that we're doing because we're creating all these leads. But it's a weird situation. It's hard to get that right. It's an aside, so I won't go into that. So what we're doing is really trying to line our content strategy up with our funnel at the top, really help on the big picture. And then as you get towards the bottom, start solving their problem with their data and then give them a plan. Before they even bought, make sure they have a plan. And what we try to do is turn our salespeople into order takers. Not to say that all we care about is transacting business with our customers, but the salesperson, that should be their job. And the more marketing educates, then you can actually make these SDRs work. You can make this inside sales work as long as you're not relying on that person to educate the CMO or the director of marketing at one of your clients. Why the heck would a 23-year-old be able to do that? Marketing can though. And turn your salespeople into order takers. That doesn't mean that you don't care about your customers. And so education is key for us. The next thing then, this is frontier now for us, is personalization. To give you a sense, we'll have 350 million people visit our website this year, different unique people. Good God, we don't know who they are. And then you blast them out with just one message. It doesn't work very well. And for us, where this manifests itself is a lot of SaaS marketing, it's all about that pricing page. We probably all love that. You just go to the pricing page, you can figure out everything. What the features are, the price points, it's great for competitive research. So we've taken this to an extreme level. This is our pricing page. I haven't figured it out yet. I've been here 18 months. And so what we've done, it's addictive. Which is, oh, just keep putting more and more on this pricing page. I think we had 34 million people visit this last year. And so we keep having more people visit this. But it's really fascinating. We can change our revenue by two or three million dollars next month by changing a few of these words. And then you just keep doing that. Oh, let's keep trying that. But what you're doing is you're actually just choosing who to sell to. You're not selling to more people. Oh, I'm going to sell more to that person. But you're always just optimizing. And so one of the things we're going to do is we're going to get rid of the pricing page. No more pricing pages. What we're going to do is really focus on persona-based entry into our web funnel. Who are you? Establish the value prop for that person. Help present them a plan. Get them in the first experience. And then get into pricing and things like that. And I've spent a lot of time with the folks that seem over a drop box and people like that. And the more and the more and the more we get rid of pricing pages, it seems to be working for everybody. So I would encourage you, if you've got a SaaS, if you're a SaaS marketing, you've got a pricing page, get rid of it as a test. See what happens. You'll probably end up with people more qualified coming into your funnel. Another thing that we, as Prezi, if you know us as a company, we've been shifting more into business use cases because we're just finding our product there. And so as we go more to enterprise sales, this idea of account-based marketing, where you have very targeted marketing for specific named accounts, that's another area of personalization we plan to go into. This is a basic statement, support your customers. But the point here, and I think you guys all know this, the math of SaaS, if you have a strong upselling customer base that's renewing your gold, right? And if you don't have that, you're going to fail. And if you follow this investment bank, Pacific Crest, they're in San Francisco, they study SaaS really well. I would encourage you to. This chart, which is hard to read here, the green bars, the good thing is they're showing that outperforming companies, so companies that are outperforming in the public markets or even small companies, startups that have less revenue at the bottom, they do 50% better in their upsells. It's all about upselling your existing customer base. The performance of a SaaS business is all about upselling your customers. And the good news is it's six times cheaper to upsell a customer than to acquire a new one. But I think we sometimes forget about that. Or we think about it on day 350 of the renewal cycle. What I would encourage you to do is think about marketing and that support function. So we're in the process of launching, let's say, like 30 how-to videos that are 30 seconds short. And we're going to pay, actually. We're going to pay to put this into the channels where our customers are so they get trained. Why wouldn't you pay money to train people who have already started to engage with your product? It's so hard to get them to do that. So I would encourage marketing to really think your job is also customer success. And then lastly, through all this complexity, I would encourage all of us to always think that at the end of the day we're doing this to get our product into the hands of someone with a need. So keep remembering as we hire the bots and as we tweet and we do all that stuff, it's really about getting a product into the hands of someone with a need. And as opposed to sort of product building it and we tweet it, I encourage us to really think about it as an integrated customer experience. And the more friction you take out of the customer experience or those pinches in your funnel, the more your business is going to perform. And so we think of usability not in our design. We think of usability for the whole business. What's the usability of someone learning about us all the way through to using the product and renewing? It's not easy, but the more you do that, I think the more success we'll have as SaaS marketers. And so I'd leave you with, in my opinion, I think the CMOs of the future really are the people that will own the whole end-to-end user experience of the entire business. As opposed to product kind of dictating user experience, I really would encourage marketers to think about this. Because if you think about the experience of the entire business, you'll have the social login before you send a tweet and your customer support team will be ready and your whole sales team will have awesome LinkedIn profiles and be tweeting the right content. And if we line all that stuff up, I think that's sort of where this SaaS marketing is evolving. So I think I'm on time. And if there is a question or two, I'd be happy to try to field those.