Marketing Your Startup: A Billion-Dollar Company’s First Marketer Reflects Back

Ellie Mirman was the first marketer hired by the CMO of HubSpot, the Boston-based marketing software startup that IPOed in 2014. Below, she shares lessons learned from the earliest days of marketing the company and how this has translated to her second startup role as VP of Marketing at Toast.

NextView Ventures: Thinking back to before anyone knew HubSpot or the company was headed towards an IPO, where did you even start to market the company? When you were flat-footed, what did you to gain initial traction?

Ellie Mirman: When I came in, Mike [Volpe, HubSpot’s CMO] had a to-do list of the things he wanted us to do, the top one being to start a webinar series. That was our first stab at some kind of gated content that was educational and meaty enough to get people to convert on our site on a lead-gen form.

Initially, we promoted it to a list that we’d already developed through Website Grader, the free app our CTO Dharmesh Shah had built before we even had a product. So we had those two assets: Website Grader and the resulting email list.

NVV: That’s not in any book about growing a startup — “Build an interactive app to start your marketing, then host a webinar for its users.” What were the advantages of doing these aytpical, non-scalable things?

EM: It definitely helped us get off to a very fast start. Our first webinar for example got 300 people in attendance, which was absurdly great to us given that nobody knew who we were at that point. We were this tiny startup with a handful of people. But even with that webinar’s success, we still had a lot of education to do in order to get people to understand what we did and even how we saw marketing in general, which we needed to do to generate demand and sales.

NVV: Lots of companies have products with use cases, features, or general topics they could address in their marketing, so they struggle to pick just one and get initial momentum. This is even harder when you’re not talking about the product but are creating content to educate and attract people. How did you address that?

EM: We had waves of topics. We started with a general webinar about how to turn your website into a marketing machine, then we went on to SEO, then blogging, then social media. We tried to jump on whatever topic marketers were trying to learn about next. We hit a challenge once we got to social media, because it was becoming much more crowded in our space with more businesses educating and creating more content.

NVV: How did you adjust?

EM: We experimented and found success through another tactic aside from webinars on social media. Around the same time we were doing those webinars, we launched our first ebook, and that did much better in terms of conversions — click-through-rate on emails, conversions on landing pages, and so on. So we started doing both webinars and ebooks, then ultimately did more ebooks than webinars because they performed better. Webinars have a whole other cost in terms of time, effort, headaches, etc.

NVV: So that’s the stuff that actually worked when you tried it. What about some of the uglier stuff that doesn’t feel like a nice, neat marketing funnel and perhaps didn’t work?

EM: We were up for trying anything, and that even included outbound tactics like direct mail. I was incredibly reluctant about it, but it speaks to how much we were committed to testing and measuring. We would give things a few different tries, whether it was trying a topic (like social, SEO, etc.) or by trying different tactics (direct mail, webinars, ebooks, and so on).

So when we tried direct mail, it failed the first time. We tried a couple more times to be sure it wasn’t the execution that was the issue, and we used different topics each time. Then we looked at the data and made a decision to cut it.

Then there were also certain social channels that worked for awhile but didn’t stick. In the early days, StumbleUpon was a key part of the playbook. But HubSpot doesn’t even talk about that now. For awhile, that channel was a share option on all our blog posts, and we tried paying for stumbles for some campaigns. Then we realized it was driving tons of traffic that was highly unqualified, and it would just bounce immediately. Those visitors wanted to browse, not get deeper into our site and content.

NVV: Lots of marketers are under pressure to come in and generate demand right away, whether because sales is asking for leads or founders are demanding PR or user growth. Did you feel that pressure to deliver immediately?

EM: There wasn’t as much urgency right away, partially because we were figuring things out, and partially because we surprised ourselves with how well things worked. We didn’t know what to expect with our first webinar, for instance, but we wanted to get 20 people. We got 300. So our early goals were to continue to grow significantly month over month. The sales team was still figuring out who we should sell to and how, so they would call everybody in an effort to figure out how the sales process worked, who would close, and who would go on to become a happy customer. We had more than enough leads to feed that process, so there was less pressure at the start.

But marketing should always be ahead of sales to alleviate pressure. Interestingly, HubSpot isn’t unique in this respect. I’m at Toast now and see something similar. There are goals set for marketing, but our sales team also relies on more traditional tactics like networking, cold calling, going to events, getting word-of-mouth referrals and those types of leads, so they’re not as dependent on marketing yet.

NVV: Some startup marketers think about crafting a nice, neat funnel, from a prospect’s first touch all the way down to sale. Others are scattered and throw stuff against the wall to see what sticks. Where do you land on that spectrum?

EM: When you’re at an early stage, you don’t know what that buyer path looks like yet, so you should be trying all different strategies to get data and figure out the playbook.

It’s also important to match the marketer to the company. I know that inbound lead-gen is my specialty, so I looked for a startup that wanted to play to those strengths. It’s so important to line up the type of marketer and leadership that fits the business model, the product, or even the founder’s philosophy or early attempts at marketing. Some startups are in stealth mode and maybe branding and beautiful design is important. That’s great, but that’s not my specialty, so I knew I was looking for roles that allowed me to be scrappy and execute the set inbound tactics.

So to answer your question, I’m on the side of testing and experimenting with a range of marketing tactics — and the company that’s a fit for me allows me to do so to refine the funnel over time.

NVV: What about all the various tactics available to you even with the broader style of marketing fitting yourself? Even identifying as a demand-gen marketer or branding or B2B or B2C leaves a lot of decisions still on the table to actually execute. Where do you start?

EM: Well it’s so critical to focus, particularly at a startup, because you do have those limited resources — one being your time, the most precious resource that you have. So you just need to pick a way to focus from the start, then adjust based on your data.

Personally, I start by figuring out what the ultimate measure of success looks like — what’s the most important number that we want to move or thing that we want to achieve? That has to be clear at the company. You’ll have a lot of distractions if you have multiple goals and you’re not prioritizing them. Yes, customers are you number one goal, but marketing-wise, you need specific, attainable goals, whether it’s leads or more demos or free trials or press hits.

And that thing you focus on today can and should be fluid — goals and prioritizes change over time. So that’s one way to focus: goals.

Then, there are the tactics you use and what happens when you figure out one thing works. I’m a relatively risk-averse person despite my life in startups, so my approach is to try and find a few different things that will be part of the core playbook. For example, if Facebook ads are the top driver of leads, that makes me nervous. I don’t want to put all my trust just into one channel because that’s too risky. So I’ll keep that going but test out a few more strategies at the same time.

The other benefit of testing out multiple strategies is that you can suss out the false positives. Facebook might look amazing right now, but that’s just because you don’t have enough context around what other channels are capable of doing for your business. It may turn out that Facebook is the most expensive user acquisition channel for you, but you won’t know that until you test other things.


NVV: How do you not get stuck just testing and looking at data over and over? When do you start to double-down or evolve the playbook into a way of operating? Should you do that at all?

EM: My rule of thumb is to look at the length of the sales cycle. Some startups have really long sales cycles. Some have short, more transactional cycles. That translates to the amount of data you’ll get in any test and how quickly you’ll be able to see results and adapt. So at Toast, we’ve been generating leads from a new source for a few weeks, but there are questions about whether they’re turning into sales. So I want to give it a month to match what we believe to be our sales cycle.

If I had a business where there’s a six-month sales cycle instead, I’d need to wait longer and look for affects on some metrics that would provide an early indication of success or failure.

I know this answer is about lead-gen and sales-based conversions, but I think this applies to branding and PR and other types of broad marketing too. If I were to focus on PR for a month, for example, I’d want to see some initial traction to figure out what’s working and what’s not.

The other thing about this testing and learning approach is to get into the mode of “I want to learn what works” versus “I know what’s best.” Marketers can get very attached to their ideas if they equate their value to whether or not an idea is successful. And that’s a dangerous place to be. There could be a strategy that does wonders at one company but not in another startup, and that has nothing to do with that person.

So it’s less about I have the right answers to everything versus having the right mentality to approach everything.

NVV: You’ve been Frontline Marketer #1 before, and now you’re on the hiring side as a VP. Is there a certain set of traits or even an ideal profile of a great early-stage startup marketer? 

EM: If you’re a founder who doesn’t know marketing or care to pay attention to it, you don’t want to hire a junior individual contributor at first because they’ll want direction — not need it, necessarily, but want it. Marketing has changed so much and so quickly, and it’s still changing, so it can be tough to keep up for that type of hire. And some marketers don’t get a lot of time with execs because they’re so junior, so they guess at what to do and feel marginalized. So it’s important to have someone that’s able to set the direction and is involved in the company overall.

NVV: Anything else you can impart to others who may be marketing a startup but haven’t had that initial learning period like you did?

EM: That’s a tough one. I’d say this: It’s so interesting going through this a second time, which is what I wanted. I forgot about the dark times — all the times of uncertainty about whether or not you’ll get past the current obstacle or challenge.

Startups can be very emotional, and it’s important to remind yourself that the emotion you feel now — whether the highs or the lows — will probably be nothing in comparison to what you feel a year or two years from now. If you’re saying that this month was your biggest month ever, the goal is to make sure that, six months from now, it was a blip on the chart.

You can reach Ellie on Twitter @ellieeille