Product Strategy is hard. For most tech companies their first big product success is also their last. Finding a second hit is very hard no matter how much we invest in vision and strategy development. The core business somehow always generates 90%+ of revenue and engagement.
But some companies are different. Consider these magical product sequences:
- Apple: iPod → iPhone → iPad
- Amazon: Book store → Everything store → Marketplace
- Netflix: Mail-order DVDs → Video streaming → Originals
- Google: Search → AdWords → Gmail → Chrome → Android
- Tesla: Roadster → Model S → Model X → Model 3
How do they do it?
Here’s is the common-wisdom answer: a visionary leader + flawless strategy + a culture of innovation (etc.). It’s a fantastic story that the companies and their leaders are only too happy to perpetuate.
It Ain’t Necessarily So
If you dig deeper, though, you find that the backstory is far more convoluted and messy than we were made to believe. Luck is always a major component in success, and the progression is is rarely (if ever) Vision → Strategy → Product. if you’ve been following my articles you’ll recognize the alluring siren song of Plan-and-Execute.
Great vision and strategy are good stories written in hindsight.
Here’s a case in point — the iPhone — the most successful product of our generation. The official history is that the iPhone was the brainchild of Steve Jobs, derived form his bold post-PC vision. Jobs was famously willing to bet on a high-risk product that will usher-in a new age of computing, even at the cost of cannibalizing Apple’s thriving iPod business.
Apple is usually extremely secretive, but the true story of the iPhone had gradually been allowed to leak out. In The One Device Brian Merchant relays a detailed history of the iPhone as told by the people who built it.
Far from being a result of a premeditated strategy, the iPhone was a case of multiple strategic initiatives converging to create a perfect storm:
- Multi-touch — In 2002 a self-appointed group of Apple designers discovered nascent multi-touch technology and spent a year to improve it. After some convincing, Steve Jobs agreed to create a multi-touch tablet project codenamed Q79, but a year later the project was canceled due to cost. For awhile that looked like the end of multi-touch at Apple.
- Motorola Rockr e1 — In 2004 Apple partnered with Motorola to create a phone with limited iPod functionality — connect to the iTunes store and download up to 50 songs. This was purely a defensive move meant to get people to open iTunes accounts and buy iPods for the full experience. The Motorola Rockr e1, launched in 2005, but it was slow and clunky, and did not sell well. Apple was unhappy with having limited control over the final product and aborted the strategy.
- AT&T partnership — There was one upside to the Rockr project — through it Steve Jobs was able to forge a relationship with AT&T wireless that lead to a unique agreement — AT&T was willing to give Apple free hand in designing its future phone in exchange for exclusivity. This removed a major obstacle for the Apple phone. Up to this point Jobs didn’t want to make one, fearing that carriers would meddle with the design (as they were accustomed to doing and still do). With the AT&T agreement in place Apple was free to build the iPhone as it saw fit.
- iPod phone — Sensing the shift towards phones, the iPod team started a stealth project to create an iPod-based phone. The team demoed the prototype, a gutted iPod fitted with radio transceivers, a modem, and a click-wheel interface, to Steve Jobs in 2004. Jobs didn’t love it, but saw enough potential to create a project, codenamed P1 (the names were internationally meaningless to keep them obscure from the rest of Apple).
- Multitouch phone — Now determined to develop an Apple phone, Jobs challenged the ex-Q79 design team to design a multi-touch phone. After an arduous design marathon the team was able to show a demo that the CEO liked, which turned into a project, unsurprisingly named P2. Jobs then pitted projects P1 and P2 against each other in a race to build the better phone. Fierce internal rivalry ensued. Eventually the richer and easier-to-use multi-touch interface prevailed. P1 was canceled and the team switched to developing the hardware for P2, leveraging what they have learned. In true Apple style, the P1 team was not allowed to see the secretive multi-touch UI until a very late stage.
- The iPhone — The software team behind project P2 sensed a chance to create more than just a phone with multitouch. After long debates they convinced Jobs to let them equip the device with a miniaturized version of Mac-OS-X, later known as iOS. That turned the phone into a mobile computing device capable of running rich apps and doing multitasking. P2 became an ambitious and far reaching project — the Internet + dozens of apps in your pocket. It may be here that Apple’s post-PC vision was born.
Apple found the vision and strategy for the iPhone through trial and error, and the process was as much bottom-up as top-down. Steve Jobs was not the all-knowing, all-seeing inventor. He was an opportunistic leader willing to test strategic ideas, shut down the ones that didn’t work and push further the ones that that do. He connected the dots and created a coherent product and business out of disparate projects. He absolutely the title “the ”father of the iPhone”, just not in the way we think.
Multiple Strategic Tracks (MuST)
“People think innovation is just having a good idea, but a lot it is just moving quickly and trying a lot of new things.” — Mark Zuckerberg
Whether intentionally or not, great companies arrive at their product strategies through parallel explorations which I’ll call here strategic tracks. This principle is so important that it deserves its own acronym — MuST.
Strategic tracks test big ideas — new products, new business models, etc, by building short/medium size projects. Each strategic track can be thought of as a series of experiment on a grand scale. The goal is as much discovery as validation — we start with an idea and see where it takes us.
These are the strategic tracks that lead to the iPhone.
As you can see the tracks are overlapping, sometimes even mutually exclusive. That’s perfectly fine. We don’t want to force the entire org to conform to just one, unproven strategy, that would be foolish.
Testing multiple strategic ideas improves the odds of finding the next big thing.
MuST is the opposite of classic plan-and-execute strategy development:
- It’s not a about information gathering, analysis and careful planning and budgeting.
- It doesn’t necessarily follow one overarching, immutable vision. The vision evolves as we test different strategic ideas.
- It’s not conceived in an executive boardroom and handed out down to the troops for execution. With MuST strategy can emerge organically bottom-up as well.
- It’s not the brainchild of one, super-visionary, person.
Let’s look at a few famous examples of use of MuST that demonstrate why it is so powerful.
Google’s Early Product Strategy — Creating Optionality with MuST
“A strategy with optionality is like a highway with multiple exits”. — Nassim Taleb
Here’s a rough timeline of Google’s product launches in its first few years.
Google started out with Search and advertising. Then, somewhere in 2000–2001, it began branching out in multiple directions. The diagram shows only the most prominent products launched or acquired, but there were hundreds of other projects. Did Google lose its bearings?
Thing make a lot more sense through the perspective of strategic tracks.
Google’s product strategy was to build Web apps that will get more data and more people onto the Web. While many of the early projects weren’t game-changers, they opened new strategic directions. Google Groups was a good prelude to Gmail. Froogle turned into Product Search, which is now Google Shopping. The unsuccessful Google Video lead to the 2006 acquisition of YouTube. Years later these are core products of Google that make it billions in revenue.
Google made heavy use of the principle of optionality. Each track gave Google technology, code, inventions, market footprint, learnings, customer and partner relationships, at a relatively small cost. By 2005 Google had the option to enter Messaging, Geo, Media, Social and more. Some of these options it exercised, while others, like social networking, started by Orkut, it passed on. Through this process Google was able to diversify itself and avoid becoming a one-hit-wonder.
Microsoft’s — Capitalizing on Strategic Opportunities with MuST
“Success = talent + luck;
Great success = a little more talent + a lot of luck” — Daniel Kahneman
In late 1980 Microsoft founders Bill Gates and Paul Allen were presented with a dilemma. Their biggest customer, IBM, had asked them to develop an operating system for its future personal computer, the IBM PC. Microsoft had never developed an operating system, and actually Gates and Allen believed that Unix will be the future PC OS. What were they to do? Microsoft, still in many senses a startup back then, could have rejected the offer and focused on its core products — programming languages and dev tools. Alternatively it could have pivoted and gone all-in on the operating system project. But Gates and Allen chose a third option — they acquired the rights to a rudimentary operating system aptly named QDOS (quick and dirty operating system) and within a year evolved it into PC-DOS which they sold to IBM. Hidden in the agreement was a clause that allowed Microsoft to resell the software, and resell it did.In the following years dozens of IBM-PC clone companies emerged and they all licensed their operating system from Microsoft. Microsoft’s had found its big breakthrough by turning an opportunity into a strategic track.
Luck plays a much larger part in success than we think. Being in the right place at the right time is invaluable. The question is what do you do when opportunity knocks. Classic strategy theory will have you go through exhaustive research, analysis, deliberation and planning. Alas, no amount of research and deliberation could have predicted the impact of the IBM PC or of multi-touch. With MuST it’s a matter of opening a new strategic track and seeing where it leads.
Netflix — Systemizing MuST
Netflix makes the most rigorous use of MuST. In this article former Netflix VP of product, Gibson Biddle, explains how the company creates “swimlanes” to test strategic hypotheses. Each swimlane has a team testing one strategic idea by building a series of tactical projects and experiments and measuring them against predefined metrics. At any given point Netflix has between 2 and 10 independent swimlanes.
Here’s what Netflix’s strategic tracks might have looked like in the mid-2000s (right), when the company was trying to invent its streaming services, and what these end up in (left):
As an example, the “margin enhancement” track tested different monetization options, including subscription, advertising and pay-per-view. The team discovered through experiments that the best model was an all-you-can-eat streaming subscription for ~$10/month — basically what we have today.
Periodically Netflix management adjusts the swimlanes: close ones that didn’t lead to meaningful results (e.g Community and Exclusives), open others, and adjust the funding for existing swimlanes per their results. This is not unlike what Steve Jobs did at with the iPhone, but driven by data rather than product intuition. Netflix’s use of MuST paid off — at 20+ years old it’s a household name, enjoying revenue growth of ~40% YoY. With its current experimentation Netflix is likely able to test strategies even quicker, as it already demonstrated with its next breakthrough — Netflix Originals.
The Anti-pattern of Product Strategy — Big-bet projects
So high-performance companies have used Multiple Strategic Tracks to great effect for decades, albeit with different approaches.
Then, in the late 2000s, this happened:
The phenomenal success of the iPhone and the iPad made everyone stop and question what they were doing. Every company wanted to be the next Apple and every leader wanted to be the next Steve Jobs. Everyone wanted to copy Apple’s playbook. Ironically the wrong lessons were learned. We bought the PR story: a visionary CEO, top-down strategy, moonshot projects.All of a sudden everyone talked of “going all-in”, “making big bets” and “moonshot projects”. Caution was thrown to the wind and strategic hypotheses were turned into white elephant projects with little or no validation. Dogma and hype replaced MuST. The result was a series of big, expensive failures: Google+, Google Glass, Facebook VR, Facebook Chatbots, Amazon Fire phone and others. It was quite depressing to watch.
Multiple Strategic Tracks is an extremely powerful product strategy pattern that is not going away. With the adoption of evidence-based development (Lean Startup, Design Thinking etc) more and more companies are rediscovering it. With today’s rapid development, quick iteration method, MuST is no longer the prerogative of large companies. In fact any company that already found its first success should already be laying out strategic tracks for the next big thing. I’ve even seen forward-looking startups do this successfully. If this is your situation you should give MuST a go — you never know when your first success will run out of steam.
To receive posts like these by email sign up to my newsletter.