SparkToro’s Year One Retrospective

On April 23rd, 2020, at what felt like the worst time to start a new business in a hundred years, SparkToro launched for the first time. The world’s attention was anywhere but on a new software product, yet Casey and I felt we didn’t have much of a choice. We needed revenue. We needed to learn from real customers. And, we hoped, the online world needed some of what SparkToro could offer: never before available data, at scale, about almost any describable, online audience.

How Did Our First Year Go?

Let’s put it this way: if we were a venture-backed company, the last year would have just barely met an investor’s minimum expectations (and they’d probably be pissed that we’re saving money vs. re-investing in a faster growth rate). Thankfully, SparkToro has no institutional capital, and so, considering the conditions under which we launched, on the founder-satisfaction scale, we’re at an 8/10. As Casey told me last week, “when I look at our monthly financials, I feel proud of us.” Me too, buddy, me too!

We’re not yet publicly sharing actual numbers, but the above MRR chart (Monthly Recurring Revenue) illustrates five interesting points:

  1. Growth rate varies wildly – this isn’t surprising in the first year of a subscription business, as even small fluctuations in signups, churn, promotions, etc. can have massive impacts on a given month’s growth rate. One thing we’ve noticed (that’s probably unsurprising, but may help other SaaS founders): months with spikes in signups are almost always followed by months with higher-than-normal churn.
  2. Average growth is ~10% – a very respectable growth trajectory, but one neither of us anticipate long term. Our projected monthly growth rate is 4%/month, and we’ve been pleasantly surprised by the many months that have beat that.
  3. Spikes come from features & promotions – Our biggest growth month by far was September, 2020, when we introduced our most-requested feature, email contact info, and simultaneously launched the $50/month plan. Since then, new features like Press+Media sources (in January) or Text Insights (in April) have led to bumps in engagement, coverage, and new subscribers.
  4. We hit cashflow breakeven in September, 2020 – and have been profitable every month since. Our plan is to keep expenses minimal (right now, salaries and health insurance are almost all of it), expend energy listening to customers, improving the product, and focusing on an organic marketing flywheel.
  5. SparkToro is high churn – from the start, we’ve intended to be a product that’s useful to subscribe-to when you need us, easy to cancel when you don’t, and compelling to return-to when you need us again. We’ve got a contrarian view on short-term customers and SaaS churn that, long-term, I believe can help us stand out in the market. It’s why we pro-actively nudge subscribers to cancel if they’re not getting value from their account, and why we don’t have a setup fee or a sales process.

A Look at Marketing & Traffic

When I tell folks that I do almost no SEO for SparkToro, they’re stunned.

“But, but… You’re the guy who started Moz?! You’re how I learned SEO!” they’ll say.

I still like SEO as a potential marketing channel for many businesses, but it’s not how I believe we can best build our early flywheel. That’s primarily because the product we’ve built and the problem it helps solve isn’t something people search for in Google.

When I ran Moz, there were thousands of search keywords that, if someone typed them into the Google search box and ended up on Moz’s website, yielded a likely customer for the product. At SparkToro, that’s just not true. The audience research SparkToro provides is something most marketers, even those for whom the product is a perfect match, would never think to search for or even know how to describe.

That’s probably been the hardest part of this company: consistently, clearly positioning what we do and for whom.

When folks try SparkToro, they (hopefully) learn that we’ve got hard-to-find-elsewhere data about groups of people online: what they read, watch, listen-to, follow, how describe-themselves, what they talk about, etc. But, that first time you try the product out, you might not have a problem solved with that data. Or, rather, it might not be the most pressing problem you’re facing.

Our perfect customer is someone who needs our data to solve a painful problem right now. They might be (and yes, these are all real SparkToro customers I’ve spoken to) a(n):

  • SaaS founder seeking insights into their competitor’s customers
  • Market researcher looking to understand an online product’s fanbase
  • Economic development officer at a board of trade helping their government understand an industry landscape
  • Podcaster looking for ideal guests
  • Marketer at a challenger CPG brand looking for ways to outmaneuver the behemoths in their space
  • Event organizer in need of just the right speakers for a webinar
  • Search marketer looking to get an edge in their keyword research
  • Agency owner looking to win pitches with data they won’t get anywhere else
  • Consultant helping a TV series prove the value of their audience to potential networks (oddly enough, I talked to two separate folks in this world in the last ten days!)
  • Ad buyer seeking tricks to get at the right Facebook, Instagram, YouTube, and/or Google Display audiences
  • PR professional working to ID sources of influence outside the classic media databases
  • Link builder looking for unique opportunities
  • Comms executive wanting to append audience data on top of recent brand-coverage sources

Every time someone subscribes to SparkToro, I visit their profile, check out their website, and send a personalized welcome note. I *think* our largest group of distinct customers are probably strategic marketing agencies, and yet I’d guess they make up less than 10% of our total unique subscribers. It’s a great problem to have, but it certainly makes writing conversion copy, or making an onboarding email campaign difficult.

Despite that high-quality problem, we’ve amassed nearly 40,000 free users of SparkToro in the last 12 months. How? I’ve started calling it Influence (no “r”) Marketing.

The concept’s simple: I find sources of influence (people, publications, email newsletters, podcasts, blogs, social accounts, industry news sites, webinars, YouTube channels) that folks in the worlds of marketing and market research pay attention-to. Then, I build a connection or get an introduction, and see if they’re interested in “doing something together.” Some of those pitches are as easy as an email or a DM. Some, especially outside the digital marketing world I’ve worked in the past couple decades, are more difficult. Some of those calls, podcasts, webinars, guest contributions, joint research projects, requests for a social bump, etc. bring in load of new, free signups. Many only result in a trickle. Probably 1/3rd of the time, I’ll do one and can’t prove it sent even a single person to our site.

I don’t mind the serendipity. It’s how we’re building SparkToro’s flywheel:

We rely on folks hearing about our brand, directly from me doing some marketing thing with an existing source of influence, or indirectly from secondary word-of-mouth. In the first few months after launch, it was almost entirely the former. Today, it’s 80%+ the latter.

In the first 6 months after our launch last year, we received 113K visits from Google search. ~70% of those (according to Google Search Console) came from branded searches for SparkToro, Spark Toro, Rand Fishkin, or one of our free tools. The last 6 months, search traffic rose to 193K, of which ~70% are, again, branded searches. It feels weird to almost double search traffic in six months and neither rank for any new search terms nor improve rankings. Instead, we spent effort increasing branded search demand through hard-to-measure, influence marketing.

Below is a look at new, free signups the last 11 months (since we started tracking them in GA):

Free subscribers are our #1 marketing channel, and serving them well is a big priority. Many of those 40,000 folks have told us they get a lot of value from the free version, and these folks are also who’s spreading the word. Casey and I love that. Could I go back to my old ways and pick a slew of tangentially relevant keywords, create content, and try to earn some links? Sure. Would I much rather have 100 people searching for SparkToro each day than 1,000 people clicking my content for unbranded search terms? Absolutely.

What About the Team?

Today, SparkToro is still just me and Casey. For the next year or two, I expect that will be the case. We’re both busy, but not so much that we can’t take a day off when we need, or even have a couple slow-productivity weeks.

Casey at the slopes with partner, Lindsay

Rand cooking at home for Geraldine (which has turned into a full-blown hobby)

That’s not to say we’re alone at SparkToro! We’ve embraced the outsourcing model (something I recommend to everyone), utilizing more than a dozen independent contractors, agencies, and consultants to help us with everything from classifying podcasts to conversion rate optimization to taxes, accounting, art design, UI/UX, and more. I love people. And I love working with people. But, I don’t see any need to bring them on full-time and try to find or make work rather than paying as we go and leaving everyone free to optimize their productivity and skills as they see fit.

Our bias is to get help from the best folks we can afford for a few hours when we need them most rather than hiring, onboarding, and training someone whose full-time work we could afford (i.e. not top-tier talent).

Perhaps unsurprisingly, this work environment yields remarkable productivity. I’m constantly amazed by how much we can get done in a week, and even more amazed how it usually goes: in bursts. Casey or I (OK, it’s mostly me) will have a rough few days where it feels like barely any progress is made, followed by a day where that’s just Boom, Boom, Boom. I feel extraordinarily lucky to have a cofounder, a business, and a structure that allows for that inconsistent, low-then-high productivity model.

What’s Next for SparkToro?

Our big plans are fairly modest. We hope to maintain growth, but not attempt to artificially inject it with heavy ad spend, burning cash, or a cycle of pricing promotions. We plan to keep providing high-touch levels of personal service to everyone who emails us (except for you, guest post spammers!), and to scale that support in the product by fixing things fast and making iterative improvements based on feedback.

The next few projects we know will launch are:

  • Demographics – a great number of marketers, researchers, and owners want demographics, and we’ve found some good ways to take an early stab at providing this data in the product
  • Audience Tracking – sometime in Q3 or Q4, you’ll be able to select an audience search and track that group’s behavior over time, receiving regular updates on the new sources they follow, what they start & stop talking about, how their demographic and behavioral attributes are changing, etc.
  • German & Spanish – These two languages are our most requested, and we hope to index significant quantities of profiles in both and make them available by 2022.

Beyond that, our roadmap is malleable. We’ve designed it intentionally to be opportunistic and never too overwhelmed to deal with putting out a fire or experimenting with something new.

When it comes to company performance, our big, hair, audacious goal is to repay our investors their initial sum by the end of 2022. We have a very unusual investment structure, but fingers crossed, if we can keep up the slow, steady growth, we’ll be able to prove that our weird model can work far better than the venture path. If we do, we’re hopeful that can spur more companies and investors to consider this alternative.

If this sounds like the opposite of Blitzscaling, the opposite of how tech startup wanna-be unicorns are supposed to behave, good. We don’t think very highly of the unicorn-or-die-trying model, nor of most unicorn tech companies themselves. Concentrating wealth, building monopolies or near-monopolies, consolidating markets, crushing competition… Yawn. I’m tired of that kind of capitalism. The odds suck. The people are unkind. And the difference between earning $10M and $500M isn’t much on the lifestyle-improvement front, but it sure seems tough to stay a good person at those bigger numbers.

Answers to Your Questions

On Twitter, I mentioned that I’d be writing this post and invited folks to ask questions. I’ve tried to select some of the most interesting ones that went unaddressed by the above for this section below.

Scary. Uncertain. Emotionally conflicted.

It was such a mind-numbingly, overwhelming, life-altering time for everyone, and it just felt deeply uncomfortable to wake up and walk out to my garden-shed-turned-office to try and figure out ways to get people excited about audience research software. Tying the launch to a charitable donation (we gave $1 to GiveDirectly for each person who tried SparkToro in those 2 weeks) helped make it feel a bit less purely self-promotional, but it was still quite awkward.

Notably, the early access emails we started sending in late February showed a really fascinating, but brutal trend over the ~8 first weeks of the pandemic. Each time we sent a new batch, more and more emails bounced back as undeliverable because the marketer was no longer employed at the business/agency. Seeing that in my inbox every morning was a stark reminder of how scary those times were.

All that said, the months after showed just how resilient modern economies and businesses can be, even in the face of great tragedy and rough circumstances. Overall, while I think the pandemic certainly hurt our launch, I’m not sure if it’s impact was all that big on our overall long-term trajectory.

Yes, it’s certainly something we think about. However, it’s also one of the big advantages of choosing to be an indie/Zebra-style startup (vs. needing a billion dollar exit). If Twitter or Facebook or LinkedIn or whomever remove publicly accessible profiles and go entirely private, we can pivot to sourcing data from other networks (Reddit, Twitch, Tik Tok, even an open-web-based index is plausible with our model).

Quite honestly, we haven’t. That’s not because I don’t believe that values and principles can be useful, but because at such a small size, it’s simply not meaningful yet. I suspect that longer term, if/when we hire a true staff, we’ll be looking to put more of those ideas into play.

Side note: The one exception to the above is our statement on racial injustice and commitments we’ve made in how we choose vendors, disapprove certain types of customers, and collect/show data. That’s obviously a principles > finances decision, and one we continually consider and invest in.

Definitely. I’d say it happened for me three times: once during those first few weeks of launch, when I thought the pandemic might brutalize economic opportunity and marketing investments so deeply that we’d be up a creek before we could start. Second: over the summer of 2020, I had some unfounded-in-data fears about whether we could get growth up, fast enough, to get to profitability. And Third: the last couple weeks of December, I was in a really bad mental place after my grandmother died and we couldn’t travel to see family for the holidays. Honestly, all three of those were less about actual numbers and more about how I *felt*. Managing your own psychology is definitely one of the hardest parts of building a new venture.

Biggest motivators for us is half what our customers/audience tell us they want/need (like email contact data) and half what we think will move the field forward and create “never-thought-to-ask-for-that” kinds of value (like % of audience following).

As for difficult decisions, I wrote a couple weeks ago about the strange bias we’ve had against emailing people too much, and how we’ve had to overcome our fear of regular outreach to build a better, healthier marketing engine.

Really tough to know for sure! We haven’t A/B tested any of those things, but my suspicion is that it makes signing up feel far less risky, and creates a lot of customer and community goodwill. Those are hard things to measure, but in a way, it’s less about maximizing our financial outcome and more about what we want to put out into the world. Strange for startup founders in a brutally capitalist, winner-take-most country to say, I know, but that’s who we are (pretty sure you’re that way, too, Thad, something I love about you my friend).

  1. Robust email list development
  2. Freemium vs. Free trial (we chose freemium for SparkToro, which I think has been a big win so far)
  3. Getting good at iterative, quick launches of smaller features one at a time, vs. big, hairy launches
  4. Uptime and customer responsiveness on the engineering/tech side (that’s all Casey, but Moz struggled with it for literally the entire time I was there)
  5. Using agencies and consultants for everything (something I wish I’d done at Moz)

It’s weird getting older, because the sharp edges feel sanded down. The highest highs I’ve ever had were definitely in those early years at Moz. Being a young, 20-something, everything felt so new and exciting. But, on the flipside, any misstep or downturn was devastating.

At SparkToro, I’m certainly more fulfilled, happier with what we’re building and how, more confident than things will go wrong, and that it will be OK when they do because we’ll address the problems as they arise. Even if everything goes horribly and we have to close the business entirely or completely change what we do, I feel good about what we have done, our plans, our intent, and that people will give us the benefit of the doubt (our investors included). There’s a deeper sense of calm, thoughtful progression vs. the rambunctious excitement of those early years at Moz. In a way, I miss it. Those elated moments of feeling like you might be onto something in your youth are addictive.

Maybe it can’t quite be compared… They are two very different kinds of enjoyment. It’s like comparing your 1st date with a potential romantic partner when you’re young to your 1,000th date with your long-term spouse when you’re older. Both good, but very different.

It’s so hard to say! If I didn’t have a brand in the marketing world, I’m not sure I would have started SparkToro at all. I really believe there’s a power to founder:market fit, and that if you’re trying to build a new concept in a niche field with lots of established players, thousands of tools, software budgets that are already stretched thin, tons of very successful content marketers already vying for every ounce of attention… Yeah, I would have attempted something else.

SparkToro, and the entire concept of passively collecting scalable market research through public profiles, is super exciting and interesting to me. Could you build a company like that without deep knowledge of the digital marketing world? Without connections to people in that space? Without a few years of building up a marketing flywheel that earns ongoing attention? I’m not so sure.

I did do some trend metrics, but yeah, we’re generally not “building in public,” so to speak. That’s for a few reasons:

  1. Unlike my Moz days, lots of folks in SaaS and startup world are already doing that, so the need feels less acute. When I started Moz, I was frustrated at how few examples there were of open, transparent founders and companies. Today, there’s thousands. Could we create some value by participating? Sure. Would it be as unique and uniquely valuable as it was in 2006/7? No.
  2. Moz’s public numbers inspired a lot of competition. For most of my years as CEO, I didn’t mind and barely noticed. I took as flattery when competitors’ founders/execs pointed to those numbers as inspiring them to enter the field and compete with Moz. Over time, though, I think we saw that when Moz slipped up for a few years, those competitors quickly ran away with the market. I’m sort of hoping we give ourselves a little more of a head-start with SparkToro by staying a little quieter (not that the numbers are all that exciting or inspiring right now).
  3. Casey doesn’t want to. If he did, we’d probably do it, because my bias is toward transparency for its own sake. But I respect my cofounder’s wisdom, and I don’t feel passionately enough about it to override his very valid hesitations.

All that said, if/when we pay our investors back and start being able to offer dividends, I do hope to take more of a public, sharing position on that stuff, mostly to help inspire the SparkToro funding model as an alternative to other ways of raising equity capital.

Re: SparkToro’s values… I hadn’t seen your values presentation / post before now Will, and I have to say… I want to copy it! Pretty much everything but the actual product vision is aligned with my own thoughts on the matter. Intentionally building a great place for a diverse group of people to work, focusing on customer outcomes, writing things down, enjoying the journey. I think I’m just gonna steal your excellent work 😉

SO MUCH EASIER. Everything is ridiculously easier, from managing one’s psychology to identifying problems to prioritizing work, focusing on what actually matters vs. doesn’t, not beating oneself up for bad decisions, earning higher returns on each investment made, the lot.

The only thing that’s harder is the energy. I find myself, as I age, struggling to be as relentlessly willing to pour raw hours into work vs. what I did in that first decade at Moz. Some of that is, I think, the natural slowdown of age, but I also feel like some of it is maturity and drive. My drive for millions of dollars isn’t nearly as all-consuming. I want to have a good, comfortable life, but I’m not crazy excited about being wealthier than everyone else. I also know myself better, and I know that when I’m tired, burnt-out, and putting in hour 55 or 60 in a week, my work quality is so much worse than if I do that same work in hour 10 of next week. Does that make sense?

If you’ve got more questions, feel free to leave them in the comments below and I’ll do my best to answer! Or maybe I’ll see if Casey wants to do it 😉

Thank you to everyone who’s helped make SparkToro’s first year such a unique, wonderful, affirming experience. I’ve never been happier or more fulfilled in my entire career, and once the world opens up again, and we can all travel, go out to dinner, and hug, I hope to celebrate with lots of you in person.

Happy birthday SparkToro! Here’s to many more ahead.