I work at a company that makes data pipelines and I like to tinker, so when I came across a deficiency in Garmin’s reporting, I decided to see if I could build a basic pipeline stack that would deliver the analysis I was looking for.
I’ll admit from the outset that my solution is overkill for the problem—I could have easily used vLookups in a spreadsheet to get the answer. That wasn’t the point, though. I had the itch to build something and wanted to try out some new RudderStack features and other data tools. (RudderStack is the company where I work.)
The result has been more helpful than I thought it would be and opens up many more possibilities for deeper analytics, so I wanted to write about the process both to document the project and for the off-chance that some other data-savvy Garmin users wanted to try to replicate it (and hopefully help iron out some of the bugs).
This is the story of my little pipeline project and why I built it.
We’ll start with the big news: after several years of consulting, I joined a San Francisco-based startup called RudderStack1. I’m incredibly excited about the product and team, but more on that later.
The consultancy, Yield Group, is still trucking, helping mid-market companies achieve data-driven growth. I’m also an active part of the team running OffSite, an ad tech product that we launched out of the consultancy.
I’ll share details on RudderStack later in the post, but first, a quick look back.
Being an entrepreneur with my hands in a variety of projects, I receive a fair amount of email. There have been points where hundreds of messages hit my inbox in a day and points where that number was less than a hundred, but suffice it to say, my work routine tends to involve processing large amounts of information in my inbox.
Over the years, I’ve tried almost every email app out there, from Outlook and Apple Mail to Sparrow (RIP), Postbox, Mailbox (RIP), Spark, Newton, Astro, Boxer and others. The result is always the same: I end up back in Gmail’s native interface (via Mailplane) because it’s the most intuitive, mouse/trackpad-free email processing experience available—and I there are things I certainly don’t love about Gmail.
About 6 months ago I was granted access to an email app called Superhuman2.
This is the second post in a series I’m writing on running a services business3.
Also, For the sake of this conversation, I’m discussing services in the context of people who want to grow their business significantly, as opposed to maintaining some sort of local maximum.
Many services businesses achieve stable success leveraging their personal networks and local community. As I wrote in the first post in this series, though, those sources of new business can be a local maximum for a company that wants to scale. Large, densely-packed metro areas lessen the effects of the ceiling, but those environments are generally much more competitive and noisy.
No matter where you are, though, if you want to grow a services business, there are generally two main paths (in my observation and experience, at least) companies tend to follow. (There are certainly hybrid business models, but that’s another post for another day.)
My working names for the two paths are “productization and commoditization” and “specialization and customization.” Today we will talk about productization and commoditization.
I’m writing a short series4 of posts about observations I’ve made after transitioning from running a more direct-to-consumer focused business to running a B2B services business.
For a long time now I’ve been working to build out world class data-driven marketing infrastructure and teams, and most recently I’ve assembled a team that helps other companies do the same. It’s a services business called Yield Group that helps companies collect and use the data that will help them grow.
It’s been a fascinating experience coming off the heels of running a company that was primarily direct-to-consumer with a focused set of products. One of the most interesting differences to contemplate has been how services businesses scale.
This is the fourth post in an ongoing series about the transition from maker to manager5.
I still remember the one of the first descriptions of leadership I heard from a successful business person that I instinctively felt was wrong. I was young, early in my career, and the CEO who ran the company I worked at had joined a meeting about a project assigned to my boss’ team.
I can’t remember the specifics of the project, but I do know that it had encountered some sort of difficulty and the CEO was stepping in to help get things back on track. Seeing the primary leader of the business ’get their hands dirty’ with the team, especially alongside a young, inexperienced person like me was encouraging and felt right somehow—a salve for the notion that executives use hierarchy to shield themselves from the actual work being done.
Towards the end of the meeting, though, the CEO said this:
It’s not good that I’m in this meeting. My goal is to make myself unnecessary, so you need to figure this out on your own next time.
I was recently invited to an event about digital marketing. At the beginning of the talk, I mentioned that the material included lessons learned from The Iron Yard, a company I co-founded and helped to run for over five years. (The Iron Yard was at one time the world’s largest in-person code school6.)
After the talk, an older gentleman came up to me and said, “I’d like to have a word with you.” Not sure whether to expect praise or reprimand, I said, “I’d love to talk,” and stepped aside with him. As he pulled out his phone, he told me that during the talk, he’d sent a message to his grandson letting him know that he was at a talk being given by on of the founders of The Iron Yard.
Over the past few weeks I’ve had conversations with a variety of people about the marketing/sales funnel, which I wrote about recently7. Even though many would consider a clear understanding of where your customers are coming from—and how to scale those channels—an essential part of business (which it is), many companies simply haven’t defined their marketing/sales funnel.
A sales funnel, when combined with marketing techniques such as a streamlined marketing CRM, can be an important way of ensuring your brand delivers exactly what it is that your customers are looking for.
For those who have an intimate understanding of their funnel, the temptation is to be critical of businesses who don’t. When you step back, though, the context for most companies without a funnel is understandable and, in some cases, even justifiable.
A few weeks ago I gave a talk explaining some of the lessons I’d learned about marketing as I helped grow The Iron Yard. As I reviewed my presentation with the conference organizer, one slide towards the beginning caught his attention. Here it is:
His advice on that slide was, “you should explain what the marketing funnel is. Not everyone will know what that term means.” His observation was a great reminder that, when you are fully steeped in a discipline, concepts that are second-nature to you probably aren’t to people outside of that discipline.
286 days have passed between the last post I published and this one, which is quite a long while as far as I’m concerned. In reality, though, 3/4 of a year of dormancy probably isn’t that bad relative to the general entropy facing long-form blogs, though I don’t have the statics to prove it. Either way, I’d like to think that I’m back in action.
I have many reasons to resurrect this inert writing project: I truly enjoy writing. I’ve made fascinating connections through sharing my thoughts on the web. But the most important reason surprised me: I could feel the effects of not writing.