Growing up with and in PM and “Big Data”
Sampler cfrln
Sampler
‎12-05-2016 05:20 PM
‎12-05-2016 05:20 PM

Interana’s offering free access to this gartner report on product management’s role in digital innovation and this one on establishing product management to drive digital business success. It's a great resource for anyone who is a product person or who is making the case for product in their company. Seeing this report made me reflect on my own two decade plus journey into PM and data.

 

Like most of my generation, I had a meandering journey to product management. These days you see kids graduating from top MBA programs wanting careers in PM. That wasn’t the case in the early 90s. When I got a business degree in ‘93 I had honestly never heard of product management. Along the way, I found it was my true calling and it has been an immensely rewarding career. I also happened to make my PM career in what became cringingly known as the "big data" space – the megatrend of our time. Just as much of a vocation because as a PM I am perpetually data hungry.

 

In these two decades, and particularly in the last 5 years or so, PM has definitely been on the rise. Back when I started, PM, if it existed, reported to marketing, and VPs of Marketing were ultimately responsible for telling R&D what the market wanted and getting sales enabled to sell it.

 

I had some analyst jobs early on in the mid 90s that supported product strategy reporting into marketing. When I got a VP PM job in 2003 reporting to a venture-backed company CEO instead of marketing, I was a rarity.

 

Part of why PM has risen and become independent is that the marketers have become more focused on building demand and brands, and much less technical, much less intimate with users and use cases and much less long-term strategic. PM sometimes moved over to engineering but over there it tended to lose touch with customers and the market.

 

Today, tech startups in every category are hiring more PMs earlier and get VP and C-level heads reporting directly to CEOs earlier and more often. Empowering and making them successful is something savvy board members counsel founders to do early. And as digital transformation spreads to every industry, Fortune 500 companies are starting new software-centered business units and modeling them after Silicon Valley tech startups. Which means they are hiring PMs too.

 

All this results in lot of newly minted PMs running around trying to define the job and make the best possible impact on their companies. Gartner and other analyst firms that help big companies make sense of technology are getting wind of this and starting to publish research (like the report above) on the PM role. Interana and other enterprise software companies are sprouting up to help meet product management’s unique needs -which are distinct from both marketing on the one side and engineering on the other.

 

So what are PMs’ needs? Let’s start by looking at what we do. Our job, fundamentally, is about making the best possible match between a company’s unique capabilities and the market opportunity to maximize future profits. So PMs need to figure out what their company’s unique capabilities and complete landscape of potential markets are.

 

Now let’s think about who we are that lets us do it well. We’re simultaneously analytical and synthetic thinkers. We are flesh machines for combining data with ideas to produce knowledge and insights.

 

Good PMs are perpetually knowledge-hungry. We want to know more about our company’s technology and every remotely possible market for it than anyone else in the company. Nothing makes for a worse day for a PM more than being caught not knowing something relevant about our product, our market, our users, our company, any of it.

 

That sounds like a segue into data, and it is. It so happens that PM is now co-evolving with big data and the rise of new data-focused roles.

 

But we’re not pure analysts or scientists – we have ideas. We look for unexpected insights. We have a wealth of qualitative knowledge that informs how we look at data. We aren’t afraid to let the data prove us wrong, but we don’t accept others’ perspectives on data uncritically. We want to look at the data from a million different perspectives informed by our ideas to see what story makes sense and gives us insight into new opportunities.

 

At the beginning of my tech career, in those analyst roles I often helped PMs and execs find and quantify new market opportunities. Once I had a lot more qualitative knowledge of technology and the software market and business, I became a head of product and helped build a couple companies from zero to tens and hundreds of millions a year in revenue, one of which is a multi billion-dollar public company today.

 

Every day I looked to get more data, more knowledge and develop more insights into our current and potential users, our technology and product and every other product in our ecosystem. The questions I formed got smarter the more I knew, and the more I knew the more questions I had. I was constantly looking for different angles to get a leg up on the competition. I’d make hypotheses then look for data to support or disprove them.

 

Both software companies where I ran product in the 2000s were in the same nascent log data market. At the first company the data I had myself was limited to 3rd party research, 1:1 interviews and sales reports with a shockingly small sample size due to our high end enterprise sales approach.

 

At the second company, we instrumented our software and designed our business to give us rich raw data on who visited our site, downloaded our product, installed it, logged into it, read our docs, and more – I was constantly looking at it to get a sense of what patterns were prevalent or emerging. I couldn’t see details of how the product was used beyond the login screen because it was an on premises software product running behind customer firewalls, but the data I did have gave me a lot of insights to inform my decisions. 

 

This was a whole new world than before. Marketing was interested in totals and funnels. I was interested in paths, topics, outliers, queries, frequency, running versions and other questions about behavior. Most of the time when I would dive into data I had only a vague sense of what I was looking for and meandered through it. I couldn’t have written a requirements document for the report I wanted – I wanted awareness and immersion, not a report. My meandering was guided by deep domain knowledge of our market and product. Often I got frustrated that it was too hard to use the data given tools limitations. But I persevered. This second company is the one worth billions today.

 

Today many enterprise software PMs work at companies offering a SaaS service. Their companies have much more granular data on how their customers’ employees use their software than I had in that massively successful company where data made the difference for me. Yet PMs have new challenges using that data to get the knowledge that they need. They’re often limited to accessing report and dashboard oriented BI tools with very limited dimensions, or mobile or web analytics products with a limited marketing funnel oriented set of metrics. If they want to ask a new question to test a new hypothesis, they usually have to go to a data team and wait for that data team to prioritize the request, write custom queries, and get results from long running batch jobs.

 

Yesterday’s analysts have gone and gotten some new degrees and gotten data scientist titles and gone to work in those new data teams. As big data hype and technology has been exploding, companies increasingly hope for magical outcomes from the practice of data science. The reality on the ground is that good product managers are critical to find meaningful perspectives on all this data and point data science in the right direction to have the biggest business impact.

 

The holy grail for today’s product management is to be able to explore all this rich new data independently with no limitations on the questions they can ask and fast response times that make it painless to ask a lot of questions iteratively.

 

A big part of what Interana is about is freeing PMs and others with knowledge and ideas to explore their data interactively. To me that’s the engine of innovation.