8 min read

$43M Series D Retrospective

We (Salsify) raised a $43 million D round over the summer, totaling $98 million funds raised to date.

As I did after raising our C round, instead of writing a “ra ra!” post talking about the success of the company, I wanted to look back over the time since the C round closed and think about what, in retrospect, we did and didn’t do so well.

First, some quick background for those not familiar with Salsify: Salsify is a B2B enterprise SaaS company based in Boston selling to brand manufacturers. The founders — Jason, Jeremy, and I — sat down to write our first line of code on September 4, 2012. Now nearly 6 years later, global brand manufacturers like Coca-Cola, Mars, L’Oreal and hundreds of others rely on our platform to drive the growth of their businesses by empowering them to create and distribute the quality content that powers today’s shopping experiences. We’ve raised $98 million to date from Greenspring, Matrix, Venrock, _Underscore.VC, and North Bridge, have about 250 employees, and closed 2017 at over 100% YoY ARR growth.

Mistakes

Re-reading the Series C Retrospective really brought home how different the company has become in such a short span of time. The challenges of the business have changed in kind and not just in size with the growth.

  • Not starting the process to recruit a senior leader in product marketing earlier…possibly years earlier. Not long ago, a16z had a podcast about the foundational importance of product marketing in B2B SaaS businesses. And I now fully grok it. The breadth of our market and the surface area of our platform is quite large. We sell reasonably transactional deals to smaller manufacturers and 7-figure deals to large ones, which is a much wider range than we had 2 years ago, but unfortunately we’re using largely the same product pitch, positioning, and messaging for everyone! Mumu messaging: fits everyone, looks great on no one. Even a couple of our best, most loyal customers have told us something to the effect of: “oh man, you don’t know how to talk to people like me.” With industry breadth and scale the one-size-fits-all message that’s served us so well for years is no longer sufficient and frankly hasn’t been for a while. We’ve had a couple of great product marketers (who themselves are relatively new to Salsify!) and no real leader. Honestly, I think about how much faster we could have grown in the last year if we had an effective product marketing head and personally beat myself up — I was running Marketing most of 2017! We only just hired one.
  • Not having enough buffer in customer success…and still pulling through(Note: a “Win” I mentioned last time was over-hiring for CS…) We had a killer Q4 in 2017 and brought on dozens of new accounts, many quite large. Sales was stoked. Management was stoked. The board ecstatic. Customer success: crushed. And not crushed for a basic error like failing to plan capacity effectively, being behind a hire, or anything like that. It was a confluence of committing to a couple larger-than-anticipated projects, having a bunch of customers delay onboarding so that they were stacked on top of each other, and things like that. That whole department worked their absolute butts off in an unsustainable way for months to get through that period so that this resource shortage didn’t impact our customers. We learned a lot from the period and should be in a better, more resilient position when we get a similar confluence of events again, but it’s very difficult to entirely protect against this type of thing unless you’re willing to kill your unit economics. The real win here is that we got through it without any employee or customer attrition.
  • Scaling culture definition & messaging. I no longer know the name of every employee. One thing that happens once you pass the “I can remember everyone’s name” mark is that messages have to be more focused and directed to be absorbed by everyone and repeated far, far, far more often than you’d think. That’s not because people are stupid or anything, it’s just that what you think they should be thinking about right now while you’re telling them something you think is important isn’t necessarily always what they’re focusing on at the moment (can you remember the last paragraph of this thing that you’re reading right now without looking back?). Multiply that by 100s of people and anything you say at an All Hands meeting might be absorbed by < 50% of people present. That’s just part of the scaling game — practically every startup founder I’ve talked to has made a similar observation. Related to this: we had a culture code that was too complicated for anyone to repeat and, more importantly, not defined in a way where someone that didn’t intuit it could really know what was in or out. To scale culture, we learned, we had to create more focused and directed messaging about the culture and also make sure that it’s actionable in a universally understood way with measurements and controls on the elements we care about. That is, it’s one thing to say you care about “empowerment”, it’s another for everyone to have a shared definition of what that means and be able to speak up when they feel that you’re losing your way somehow.
  • Not opening up international offices earlier. Our exec team has some Europe PTSD from Endeca ($1.1B exit to Oracle in 2011), which never really managed to nail and scale operations over there. Four members of our exec team and several other key employees had worked at Endeca, a few for over a decade, so the PTSD is widely felt. We’ve known that we have to put offices and people over there eventually, but have rationalized why “next year” has been an acceptable strategy, even when 25% of our customer base — including some of the biggest — are in Europe. We recently took on Brett Hurt of Coremetrics and Bazaarvoice fame as an advisor, and he was instrumental in pushing us to put operations on the ground over there (location details and announcement coming soon). Thankfully. In retrospect, we should have done it a long time ago, and I’m thankful to Brett for putting the pressure on us.
  • Being very intentional about defining TAM expansion strategies. We raised a really nice D round at a healthy multiple, but in the process one piece of feedback we got consistently is that our story on how we plan to grow the product and customer base beyond its current definition isn’t tight enough for public market investors. Public market investors want to have confidence that when you hit, say, $500M ARR you can still be growing bookings aggressively and haven’t plateaued or hit some kind of a TAM wall. To have that kind of growth at that scale your TAM (Total Addressable Market) has to be very large. It’s even better if the TAM is already defined by an existing very large business, especially if that business is legacy. Salesforce, for example, just had to say, “Our TAM is Siebel and then some.” Done. At Salsify we’re creating a category vs. moving an existing one to the cloud (we’re not “PIM but in the cloud” which would be a smallish market and small vision), so the TAM is more difficult to analogize than that, and we need to do a clearer job communicating it to the public markets before IPO. (To be clear, our TAM is made up of 10s of thousands of enterprise accounts so is quite large; this is primarily a communications and position problem)

Wins

Compared to last time, the Wins here reflect notable advancements of Salsify as a potentially large business, and not just developments of the company or customer base. Again, that shows a difference in kind from where we were a couple years ago. The game has advanced.

  • PXM (Product Experience Management) category launch. Creating a new category is tough — the market has to acknowledge that what you’re doing is new and start using the same language you’re using to describe what it is that’s new — but we’ve made great strides in the last couple years. The biggest signs it’s working: competitors like Akeneo, InRiver, Edgenet, PartsHub, PlumSlice, etc., are now using “Product Experience Management” to describe what they do, both Gartner and Forrester are using the phrase, and our customers have posted job reqs using the phrase. You can think of PXM as multi-channel closed-loop marketing & merchandising for brand manufacturers (if you’re a brand that idea is golden; if you’re not you may not understand it, which is fine). For those interested in how to do this, I recommend reading Play Bigger.
  • Local hiring notoriety. A couple years ago at a board meeting David Skok of Matrix told us, “you guys don’t yet have the notoriety around town that you’re deserving of.” Through October we won at least 1 “great place to work” award per month. My hat is off to our talent team for this one. We’ve started a blog just focused on culture and recruiting, built out basic marketing automation including a mailing list, hosted tons of events at the office, and generally gotten our name out there. Most of our hires come from referrals (47%) and inbounds (25%) around town, which is a testament to how well the team has done.
  • Analytics product launch. The first truly new product and revenue stream that we’ve ever launched is a Digital Shelf Analytics solution for brands. It’s a launch into an existing-but-small category occupied by Clavis, Profitero, Content Analytics, and others, but in the 18 months since launching the alpha version of our own analytics offering we’ve gone from zero customers to being one of the 2 largest analytics providers by customer count in the space. We’re now in the process of more deeply integrating the new functionality back through the entire product offering. The ability to do this over and over again in the future will help us continue a high rate of growth as we scale. For example we just launched an Enhanced Content product with similar business and market characteristics that will be a growth driver in 2019 and beyond.
  • Breaking up the product monolith. Although we had started the work of breaking up the Salsify platform into microservices, when we raised the C round our core platform was still largely one large monolith. One of the engineering team’s great achievements in the last couple of years has been to break apart the monolith, container-ize everything, and start to build meaningful new features (like analytics) as separate projects with smaller, more nimble teams. When you’re trying to find product/market fit you need to be agile, and it’s much easier to iterate quickly on a (small) monolith than across a bunch of microservices. However, once you hit product/market fit and scale having a tightly integrated monolith makes onboarding new engineering, adding new functionality, and scaling in general more difficult.
  • Good people. Our employees are good people. I don’t mean “good at their jobs”, but rather that they can generally be trusted to do the right thing. Different cultures have different cultural ticks that permeate. For example, at Endeca raw intelligence was valued over most things, including civility at times. There were lots of strong opinions and lots of people that were “right”; this lead to a helluva lot of innovation, but it was chaotic, and different groups were often at odds. At Salsify everyone largely operates with good intentions, and I’ve never been part of a team like that. People genuinely seem to like helping each other. If you ask for help, people will go out of their way to give it to you. The culture rejects assholes (we have several proof points on this). I’m proud as hell of having a team of good people. I love coming to work.

Onward

I hope you found this perspective and the details helpful. May you make different mistakes than we have and also find your own successes.