A CEO's Guide to Marketing with Dave Kellogg | SaaStr Software Community

Hi, good afternoon. Thanks for joining me today. My name is Dave Kellogg. I'm a principal at my own consulting firm. I'm the author of a blog called Kelblog. And as we announced a couple of weeks ago, I'm starting up an assignment as an EIR, an executive in residence, at Balderton Capital, a European-based VC firm. Based in London. So today's presentation is a CEO's guide to marketing. So let's jump in and talk about the founder experience. And you know, most founders, the story goes something like this. They're working one day, and they see this kind of unserved need in the marketplace. And they think, I'm going to build a product to fill that need. and they get a team together, and they build a product, and congratulations, it works, and things are going well. And many years later, they've built a successful SaaS business. Maybe on a Sunday night, they're looking at their P&L, and they notice something funny. They say, wait a minute, I'm spending twice as much on sales and marketing as I do on R&D. I thought I founded a product company. And they think, well, maybe, maybe I'm abnormal, right? Maybe those silver-tongued folks in sales talk me into spending too much money, or those crazy marketers that waste half the budget are wasting more than half. Maybe I'm abnormal. So we go to Meritech Public Comps, and we start comparing ourselves to others. And we find out, no, wait a minute, I am normal. The median enterprise SaaS company spends twice as much on sales and marketing as it does on R&D. Sure, you have folks like Datadog that only spend as much on sales and marketing as R&D, but that's literally the 90th percentile. 9 out of 10 SaaS companies spend more on sales and marketing than R&D. 50% of SaaS companies spend more than double on sales and marketing than R&D, and it only goes up from there, right? If we look at some of the names up there, kind of surprisingly, folks like Smartsheet and HubSpot are coming in at 2.2 and 2.4. Well, Salesforce, they're a marketing juggernaut. What do they spend? Well, they spend 3 times as much on sales and marketing as they do on R&D. Zoom, 4 times as much, although off really small numbers. Monday.com, the work OS, aren't they kind of a PLG company, product-led growth? Doesn't the product kind of sell itself? Well, they spend 5 times as much on sales and marketing as they do in R&D. So in my opinion, the founder experience that happens a lot is you found a product company, and one day you wake up and you're running a distribution company. And you go, what happened? And you say, it's OK. I'm going to be all right, because this sales thing, I don't know exactly how it works, but it's super measurable. Quarter performance. Customer success, that's pretty straightforward. We know what they do. We can measure it through renewals and churn, or as we talked about last year, net dollar retention. But marketing, there's this thing called marketing. And I'll tell you what's the scariest thing a CEO can hear at a QBR. The marketing session's up next, right? Now, why is that, right? I've been 25 years in the business, probably literally 100 different QBRs. Why is the marketing session always so scary? Right? Well, think about it. What's the session most likely to take more than double its allotted time slot? Marketing. Right? What's the session most likely to leave people dazed and confused at the end of it? Marketing. What's the session most likely to generate a knock-down, drag-out fight? Marketing, right? What's the session where people's careers can literally be ended? Like I've been in QBRs where after this session I said, I don't know what that was, but that can't happen again, right? We need to get a new head of marketing, literally. So the question is, why is the marketing session so scary? And the answer, particularly for product-oriented founders and executives, is first, it's unfamiliar territory. Most founders didn't grow up in marketing. Second, you get overwhelmed by a blizzard of buzzwords and acronyms. You know them as well as I do. Marketing people love them. You're buried in a sea of numbers, conversion rates, and funnels, and web traffic, and clicks. You're confused by ill-defined terminology and process. Most marketers are not that rigorous in their thinking. And you're drawn into conflict between sales and marketing, right? So this is why the marketing session is so scary. And the other reason is, maybe, just maybe, you're a little afraid to admit that you don't understand all this stuff, right? And that makes it even harder. So what do you do? Well, you do what any good CEO would do. You wait for the session to end, and then you say, with trepidation, Well, that was great. Next session. And you continue to live a life of quiet marketing desperation, right? You don't understand it. The session is terrible. And you just say it is what it is. And I'm here to help change that today. So I'm going to be your marketing guide for the rest of this session. My name is Dave. I bring a perspective that comes from 3 places. One, I've been a CMO for 10 years at companies with as little as $1 million in revenue and as big as $1 billion in revenue. So I've been a CMO across a broad range of sizes. I've been a CEO in the $0 to $100 million range, again, for combined more than 10 years. And I've been combined well more than 10 years experience serving as independent director on startup boards. So hopefully, I'm going to be able to bring you this kind of marketing-rooted but CEO vision, CEO viewpoint on marketing. And the way I'm going to do that is we're going to talk about 5 things that every CEO should know about marketing. And it's important that this is not a generic Marketing 101 course. This is not just some quick survey course on marketing. This is rooted in my experience working with startups. So it's biased towards the mistakes I see people make and the challenges that I think they face. So that's where I'm coming from. And these are the 5 things that I think CEOs and C-level execs need to understand about marketing to work effectively with marketing. Now there's two underlying themes in this. One is use market research. In this day and age, we have so much data from our systems that there's a tendency to only want to analyze that data. And back before we had these systems, we used to rely on market research, and now we almost don't. And the problem with only relying on our data is it's kind of navel-gazing, right? We're only analyzing that portion of the market that found it into our CRM. So one underlying theme, I'll give you numerous examples, is here's a question you should use market research to go figure out. The other underlying theme is keep it simple. Don't get sucked into the complexity, stay high level. So I'm going to give you some high level models that you need to come back to, that if your marketing person is kind of pulling you down in the details, you can step back and pull things right back up. So let's jump in. Here are our 5 principles. The first one is that you can reverse engineer all of marketing from buyer empathy, right? If we just think about the buyer and we think about what marketing is trying to do, we kind of figure out marketing's job. And in effect, you can derive marketing from these 2 things. If I only knew 2 things, I could derive marketing. One, the state of mind of the buyer in terms of what question they're trying to answer. And in my mind, there's two fundamental questions: why buy mine, or why buy one, right? And why buy mine is kind of exemplified by Southwest's bags fly free, right? Bags fly free answers the question, why should I buy a ticket on Southwest? It doesn't answer the question why I should go see my uncle in Toledo, right? It does say if I'm going to see my uncle in Toledo, why should I use Southwest? Because bags fly free. Right? That is a differentiation message. The other state of mind the buyer can be in is a value message, which is, "Why should I buy one at all?" And for early-stage startups, this is often what you face. You've made the first thing, and they don't know whether they should buy one. And in that case, you need to sell value. Right? Here's the value of the thing. I exemplify this with an old Campbell's Soup campaign called "Soup is Good Food." "Here's why you should buy soup, because soup is good food, it will make you healthy." Bags fly free versus soup is good food. So answer incorrectly at your peril. Right? Just say, I work with a company called Alation, they sell data catalogs, and say when you listen to the Gong calls, which I recommend that everyone do, the whole E-staff as a group should do it once a week, and you listen to Gong calls, are you hearing things like, I'm going to buy a data catalog. I called up Gartner and they said to evaluate you and competitor B and C, and I'd like to know why I should buy yours. If you're hearing that and your salesperson gets on the phone and goes, I've just been to value selling. I know that people buy solutions to business problems. Let me tell you why you should buy a data catalog. What's going to happen? The customer is going to be like, do you have stuff in your ears? I just told you I've already decided to buy a data catalog. I want you to tell me why I should buy yours, and you're telling me why I should buy a data catalog. You're not listening to me. Right? So don't get these things wrong. And the easiest way to figure out which situation you're in is listen to Gong calls. Market research can help. Also just looking at category formation. If the category hasn't formed yet, formed yet, you're probably in a why buy one mode. Hey, I know there's this thing you never heard of. Let me tell you why you should buy one. If there is a category and there's 5 vendors in it and there's a magic quadrant, you're probably in a why buy mine mode, right? And the buyer is probably trying to figure out which one to buy. So that's the first question. We can derive marketing from understanding where the question in the buyer's mind. The second is the consideration level of the purchase. And I put this one in because there's some things, two things confuse it. One is PLG and the other is lazy marketers. And the concept in marketing is simple. We have two types of purchases, low consideration and high consideration. Low consideration, you're checking out at the supermarket, you see that pack of gum, you grab it and throw it on the conveyor belt, right? That's a low consideration purchase. You didn't think much about it. Buying a car. Is a high consideration purchase, right? You might spend weeks thinking about it, comparing models, going to Consumer Reports, right? You put a lot of energy in. Now you might think that price is the only factor that drives consideration level, but it's not. How about you want to buy tires for that car, and you're going to put your whole family in that car and drive it to Lake Tahoe in the snow. Is that high consideration or low consideration? They're only $1,000. Ah, it's high consideration, right? Because you're putting your family in potential peril driving on those tires. So we need to understand which question the buyer is trying to answer, and then how much energy they're putting into getting that answer. Because lazy marketing people will tell you the following. No one has time to read these days. We should only make short content. Nothing should be longer than a paragraph. Everything should be short. And that's wrong. Because if this is a high consideration purchase, Right? If I'm betting, say I'm buying a financial planning system, and I'm going to bet the year's financial plan on it, I'm going to get all kinds of managers through the company to use it, I have to set it up, I don't care how much it costs to buy, the cost of owning it and deploying it is high. And if it fails, I look bad. That's a high consideration purchase. So we have to understand how important the purchase is to the buyer. PLG confuses this. That little trial of a collaboration system with 3 people in your department, that might be low consideration. It's free, who cares if it works, no harm done. To deploy that same software package to 1,000 people in your engineering organization, that's high consideration, right? So we need to understand first consideration level varies, and second, to quote one of my favorite traditional marketers, David Ogilvy, long copy sells. Right? And nobody says this anymore these days, but the fact of the matter is if you're talking to your buyer and it's a high consideration purchase, they will have time to read a 12-page white paper. They will have time to go to a 1-hour demo. Right? And instead, a lazy marketer who doesn't understand this will be saying, "No, the demo has to be 20 minutes. The white paper has to be a 2-page ebook." Right? So the action here that I recommend is one, make sure your marketing is thinking about this issue, produce a mix of long and short tools, and if you're an early-stage If you're an early-stage startup, write what I call the seminal white paper, which tells your story in 8 to 12 pages to your buyer. They will have time to read it if you're thinking about the right buyer. And almost nobody does this. The best one I've seen lately is from a friend of mine who runs a company called Skyflow, and I suggested this idea to him, and he wrote a banging white paper. You can get it off their website. I think it's gated. So those are the actions there. The other way to look at this is just imagine you were the buyer. Where would you go for information? What process would you use? Who would you want to speak to as references? Important point, the answer is always people like me. So the question is, what's like you mean? Your industry, your company size, your age, your seniority, your risk tolerance and profile, right? People want references who remind them of themselves. So this is a large part of the whole Crossing the Chasm book, is you can't take a technology pioneer and reference them against a pragmatist. It doesn't work. What process would you use to make the final decision, and what criteria would you make the final decision? So to me, you don't have to just imagine this. You could ask customers about it. I think if we have 60 minutes with a customer, we spend all 60 talking about their problem and our solution. I'd actually say to carve off 8 to 10 at the end to say, "That was great. Can I talk to you about how you buy this stuff? How do you think about this purchase process? It's going to make me more effective. It's going to help my company." And ask them these questions to understand how they think about the buying process, not just about buying your software. So that's our first point. We can reverse engineer marketing from buyer empathy. The second point is that marketing is all about funnels. And this to me is the only funnel you need to know. There's a lot of funnels out there. We'll talk about some of them. But this is the one I always come back to when things get confusing. Awareness, consideration, trial, purchase. Have you heard of a product? Yeah, I've heard of a Ferrari. I've heard of a Subaru. I've heard of a Buick. Have you considered buying one? No, I've not considered buying a Ferrari. Why? They're too expensive and I don't fit in them. Okay. Have you considered buying a Subaru? No, not too much. I mean, I would have. Have you tried driving a Subaru? Well, no, actually, because you know what happened? When I was making that purchase, I didn't have time to test drive every car, so I decided to test drive 3, and I tried the BMW, the Mercedes, and the Volvo. So I was aware of Subaru, I'd considered buying a Subaru, but did I try one? No. Right? And this is it. Are you aware? Have you considered buying? And those are the reasons why you may or may not consider buying. Have you tried it, right? How easy— in software example, is it easy to use your trial, right? Do I have a free download or a free trial? Does it take a lot of time to set up? Do I have to speak to a salesperson? Big impediment usually. And if I do, do they call me back promptly? Even bigger, right? So, and then finally, did you purchase? And if you didn't, did you buy a competitor? Did you just defer the whole thing and stick with the status quo? So I was looking at buying a financial planning system, I considered buying 3 vendors, I tried 2 of them, in the end I stuck with Excel. And again, going to be a continued theme, you can and should use market research to understand this funnel. You can't find this funnel in your CRM. It's not there, right, because it's a lot of people who never made it to you. All the people who never heard of you, all the people who never considered you won't be there. So that's the first funnel. Come back to that when things are confusing. If you go to a QBR, this is the funnel you're more likely to see. Marketing will present the funnel on the left, web traffic, names, responses to campaigns, leads, marketing qualified leads. Sales will present the funnel on the right, sales, SQLs. I did a blog on this, but I always call sales S1s and SQLs S2s. 2s, and I use as my definition the safe deposit box. 2 keys have to turn. If a BDR or SDR thinks it's an opportunity, they turn their key. If a quoted salesperson thinks it's an opportunity, they turn their key. So you have to turn 2 keys to get into the pipeline, which is appropriate if you think about how valuable the pipeline is. It needs to be protected by 2 keys. So sales and sequels, we talk about conversion rates from there, solution fit, demo, which I hate as a stage, but a lot of people use it, so I put it up there, vendor of choice, legal one. You're going to see these funnels. The mistake I see people make is they forget about these funnels, which is in reality you have 4 pipeline sources in your business. Marketing, which gets a lot of attention. SDR outbound, which you should be using in a named account strategy. Alliances, which can easily account for 10 to 30% of your pipeline, usually that closes faster and with a higher rate. And then sales outbound with a targeted strategy. So the first mistake I see people make is they forget to think of all 4 funnels, and they forget to allocate a target, and a target not for ARR pipeline dollars, but a target for opportunity count to each of those pipeline sources. Now to me, the best practice here is that marketing should be the quarterback of the pipeline. Because if you think about those 4 funnels, the only person they all report to is the CEO. So either you're going to be the quarterback of the pipeline, or you're going to delegate it to somebody who doesn't run all 4. I think you should delegate it to the CMO and say, I know you're only one of those 4 sources, but I want you to make sure that we start every quarter with enough pipeline coverage. So I want you forecasting pipeline coverage. I don't want to wake up on the first day of the quarter and go, oh my God, we don't have enough pipeline, right? I want you to forecast a quarter in advance or more if we're going to start next quarter with enough pipeline coverage, and I want you to work with the other 4 sources to make sure we have enough if we don't. And you're the quarterback of the pipeline. So, and in general, look for these rates. MQL to S1, 5% to 10%. S1 to S2, 60% to 80%. Stage 2 to close, 10% maybe, but 15% to 25%. Those are all— and the ranges can be much broader. These are only rough benchmarks, but these are sample ranges for these rates. Now, as soon as we talk about funnels, that leads to models. And we can build these amazing models where we have sales bookings capacity that starts with hiring, and a rep ramp, and attrition that generates a bookings capacity. Capacity number that we shaved to get the plan number for new ARR. Maybe I'll cut 20% of that off to account for expansion, so that I get a new logo ARR target. I divide that by an ASP to get the number of deals I need, 21.3. I phase shift it 3/4 to say if I need 21 deals now, I need 107 stage 2 opportunities 3/4 prior. And then I say, well, if I need 171 stage 2— 107 stage 2s, then I need 1,641 MQLs a quarter before that. We could build this awesome model. They're all over my blog. I love these models. But don't ever forget reality, which is that's not how people buy enterprise software. Flat, full stop, period, no. Right? And this is Gartner's buying model for enterprise software. It would be 15 contacts, 4 roles, nurturing people across 2 different companies, across 2 years. Well, I used to work at Company A, and I moved to Company B. Right? Enterprise software buying cycles are complex. And the question is, what do we think about all that? And the answer is pretty simple. This quote, "All models are wrong, but some are useful." Right? And that's the way I think about this. Of course that model's wrong. Should I not build it because it's wrong? In my opinion, no. You should build it anyway, but you should understand its limitations. So we need to use models as guiding abstractions. We need to use them as tools of enlightenment, not oppression. I think the number one mistake I've seen founders make is they hold people's feet to the fire on all those sales. And I think that's wrong. I think it's demotivating. I think it doesn't reflect an understanding of the fact that the model deviates from reality. People don't like it. So I think as CEO, you should hold people accountable for the two things that matter. Pipeline coverage, basically are we giving ourselves a chance to hit the number? 'Cause if we don't have enough pipeline coverage, we're not giving ourselves a chance. And then actual sales. Did we hit the number? That's it. Those are the only two things that matter. Everything else in the model is just helping to guide us. So I'm going to skip the pro tips for now, but I have two actions I suggest strongly on this, which is because often, particularly product-oriented founders are very quantitative people, they kind of fall in love with these models and they get kind of model myopia and they start using them as tools of oppression. Everybody hates the models and they forget the reality of the underlying business, because the model's so simple and so clear. The way to solve that is two things. One, quarterly win-loss analysis at your QBR, right? Hire a third-party firm to analyze wins and losses to bring reality back into the room. And second, this is done far less often, but I think even more important, a win-touch analysis. Take 3 to 4 deals you closed last quarter and have marketing present every single touch made over the last several years. And when you're done with that, you're going to be like, oh my God, that model, I mean, is useful for making the budget, but it's so far from reality, right? So it kind of keeps you humble. It keeps you grounded in reality. Let's talk about messaging now. So we're 3/5 of the way through. Messaging is a term marketers use all the time, and I want you to think of messaging simply as a structured FAQ, so a frequently asked questions document. And it's just simply standard answers to commonly asked questions. They teach journalists in journalism school that for any story, it's about 5W 2H, right? Who, what, where, when, why are the 5Ws. How and how much are the 2Hs. I add a third H for how different, because in tech, differentiation matters so much. Like, so it's not just enough to have the 5W 2H, we should also have the how is this different from the other people doing it. So your messaging should address, for a product launch, for a company announcement, it should address all those things. And that's not to say you can't tell stories as part of those answers, because I'm a big believer in storytelling, but still the story is in response to a question. Now messaging should be simple. You should be able to tell your story on a paper tablecloth. That's my test. Whenever I test people on messaging, I never do it with a deck. I do it on either a whiteboard or a tablecloth, and I start asking them questions about the company, and I want them drawing pictures on the tablecloth, because I want the messaging in their head, because it has to be simple, right? And I like messaging to be black and white, particularly when we're on offense, we want to be black and white, only we have this. When we're on defense, we want gray, right? Well, you have consolidation, well, they have consolidation, you don't. Well, actually, we just introduced a consolidation module last week, it's kind of thin, but it's out there. And when you do that, all of a sudden you've taken the other guy's black and white claim and made it gray. And now they're obliged to explain why they're better, which is hard, right? Black and white is much easier than gray. The other thing about messaging is you got to follow the rule of threes, which is the answer to every question is 3 points. This is well known. I do have my own little spin on it, which is what if you have more than 3 things to say? What do you do then? How do we build messaging that we can remember that we have more than 3 things to say? Example, this is an approximation of Alation's messaging. Alation's a leader in data intelligence, and this is not their exact messaging, but it approximates it. What makes Alation different from its competitors? Well, 3 things, glad you asked. 3 things, intelligence, collaboration, and guided navigation. I can remember that. I can let the customer pull the thread. What do you mean by intelligence? Glad you asked. I mean, we have query log ingestion, we have a behavioral analysis engine, we have expert identification. And for each of those features, I should be able to do feature benefit, right? I won't do it now for time, but when I name a feature, I should instantly be able to go, expert identification is great because it helps you identify the right people to govern your data, which means your data governance program will be successful, right? I can go feature benefit for each of those. The other fun thing, I could do the same thing for collaboration. And the same thing for guided navigation. Here's a demo. Hey, only we have guided navigation. We have an intelligent query editor that features real-time suggestions. So it helps you build your query, and in real time, it's suggesting where you should go to find the answer you want. And it's got inbuilt data governance. It will literally guide you towards things you're supposed to use and guide you away from things you're not. That's interesting. Tell me about real-time suggestions. Glad you asked. 3 things. It could suggest tables, joins, or filters. I got 27 things now. I got a 2-level ternary tree, right? I can pack 27 things into this, right? So this is the key to building simple, memorable messaging. I still— I launched Business Objects version 6 in 1996. I still remember the messaging. PDE. Why? Because it's in a ternary tree. Right? So I firmly, firmly recommend using ternary trees to structure complex messages. Finally, last thing you need to know about marketing relates to hiring your next CMO, because CMOs last 1.5, 2, 2.5 years. You're going to be hiring one soon, whether you just hired one or not. And the number one mistake I see people make in CMO searches is you get down to the last 3 candidates, which is when they call me. They say, "Hey Dave, we got 3 candidates, love you to meet all 3 of them." I say, "Great." And I meet the 3 of them and I go, "Apple, pear, orange." They're completely different. You're a demand gen person, you're a product marketing person, you're an SDR leader, or a comms person, or a brand person. Apple, pear, orange. And I say, "How is it that we ran this process and we paid recruiting firm $100,000 and we're at the end of it and we have 1 apple, 1 pear, and 1 orange?" Right? And you may not want to admit to yourself that you do this, because a lot of people are in denial, but that's wrong. If you get to that place, you screwed up. Right? Because you never, at the start of the search, said, "We need an apple for our head of marketing," or "We need a pear for our head of marketing," or "We need an orange for our head of marketing," and at the end, I had 3 oranges, and then I got to pick the best orange. Right? What do I mean by that? Marketing people come in different flavors. You can think of marketing as having 4 pillars. Product marketing, launch messaging content, demand gen, digital programs, events. Marketing comms, PR stories and brand. Sales development, inbound, outbound, hybrid. Note the rule of 3s apply to the pillars, by the way. So in any case, marketing pillar, well, there's 4 pillars, but sometimes there's only 3. Sometimes SDRs work in sales, but at least each pillar has 3 subpoints. And the issue here is marketing people, like, They don't make 5-5 anything. There's nobody who's a 5-5. God doesn't make them, right? You either grew up in product marketing or you grew up in demand generation, right? You also don't get 20 pointers. I think 15 is the maximum number of points you can get on this scale. So the first thing you have to do is be realistic and say, what does the market have to offer me? And you only get 15 points across the 4 pillars, and then you need to decide what you want and then go Tell the search firm, and then find 3 of them. And that's the way to avoid ending up with a search that results in you on the last day of the search deciding whether you need somebody stronger at demand gen or product marketing, which is really not a good place to be. You don't want to be there. Personally, I'm a 5'3", 5'2". I think it's very hard— not only— I mean, God doesn't make 5'5"s. I'm not sure he or she made many 5'4"s or 4'5"s, right? Those first 2 pillars, it tends to be 5'3" or 3'5". There aren't many people who can honestly say, I'm a 4'5". 4 or 4.5. So think about the pillars. You can map your candidates out like this. You can just put the 4 pillars up there and put each candidate's score. And in this case, I think we need to restart the search, because the first bar is what we're looking for, and nobody really approximates it. So those are pillar profiles. So to wrap this thing up, odds are you're spending double on sales and marketing as you are on R&D. Welcome to running a distribution business. Distribution is about sales success and marketing. Marketing does not need to strike fear into the hearts of you and the E-team. By the way, great marketing people make it easy. They love to teach marketing. Bad ones obfuscate. And if you remember these 5 things, it'll be much easier for you to work with your marketers going forward. First, reverse engineer marketing for buyer empathy. Second, marketing is all about funnels. Third, marketing— use models as guiding abstractions. Fourth, message 4, messaging is a structured fact. And 5, use this notion of pillars to hire your next CMO. So thanks very much. I'll post the slides on my blog. And thanks for attending the session. Take—