Last month, I was invited to speak to the Society for Information Management – Advanced Practices Council (SIM-APC) by Professor Lynda Applegate, one of my teachers at Harvard Business School.  SIM-APC is an exclusive forum of 40 CIO’s who value directing and applying pragmatic research; exploring emerging IT issues in-depth; and hearing different, global perspectives from colleagues in other industries.

The session, “Emerging Information Technologies and Business Innovation” was geared towards helping CIO’s think about which new technologies to bet on, and best practices for applying new technologies to business problems.   The invitation gave me a chance to reflect on, and speak about, what I’ve learned in my 20+ years in technology as a developer, business user, venture capitalist, and entrepreneur.  It was a ton of fun to talk to these super heavy-hitting tech exec’s about innovation – one of my passions.

My talk addressed three questions.

1) how should a customer think about picking between an established vendor and a new start-up in a given technology area?

2) How can a customer (or partner, or investor) judge the likelihood that a fledging company will thrive?   What are the key success factors for a new startup?

3) What are the “best practices” in building and managing a new technology company?

 In this note, I focus on the first question…incumbents vs. start-ups.

Sustaining vs. Disruptive Innovations

Purchasers of new software face a difficult set of questions when a new technology (like multicore software) emerges.  Should the CIO, VP Engineering, or whoever bet on an upstart new vendor, or stick with an existing, incumbent vendor?  The answer lies in considering the type of innovation that is driving the opportunity.    I believe established vendors are best for “sustaining innovations”.   New companies are best for “disruptive innovations.”  I’ll explain. (I am borrowing from the ideas of Prof. Clay Christensen.  His book “Innovator’s Dilemma” is one of my favorite books on innovation.)

Let us consider a typical scenario.   A new product category arises in the market – propelled by a change in technology (e.g., multicore computing) or a new business requirement (e.g., Sarbanes-Oxley).   In this typical scenario, at least one new company is launched to serve the new need.   Lets call this new company HOT (Hyper-charged Original Technology).    In the other corner is SAFE, Inc. (Successful And Financially Established Incumbent), a large and well-established company.   When should you bet on HOT?   When should you play it SAFE?

I’ll start by acknowledging that, while I have a bias towards startups (the bulk of my career has been spent helping to launch new products and companies), SAFE has many advantages.  Most obviously, SAFE is, well, safe.   Usually, you do not need to worry about whether SAFE will be in business long enough to support its product after you install it.   SAFE has ample resources to provide training, service and support to help make deployment successful (assuming they can execute).   Often, new software products need to integrate with other pieces of software already in place.   Who better to deliver a well-integrated offering than the vendor of the legacy software?   Perhaps SAFE is already one of your vendors.   You know the sales and support people, they’re already in your billing system and on the approved vendor list.   And you know that no one ever got fired for buying from SAFE.   Why would you ever consider anything other than SAFE?

Some would argue that large companies don’t really innovate – but is that really true?   I found some interesting data in an article from CIO Insight.   The top 81 US-based software/internet/computer companies invested $47 billion in 2006 – with giants like Google, Intel, Microsoft, HP and IBM averaging a whopping $4.2 billion.    By comparison, total venture capital software investments in Q1 2008 were at an annual rate of $5 billion (source: NVCA).   Much of that VC investment gets spent on things other than R&D – sales, marketing, and other overhead like CEO’s.   If we make a rough guess that 50% of the VC money goes to R&D, that’s $2.5 billion per year.

Let’s get this straight – Google and Microsoft alone outspend the entire population of venture-backed software companies by a factor of three.   Either someone is wasting a lot of money, or there is in fact innovation happening at big companies.   In fact, the large vendors have a massive R&D resource advantage.

SAFE: Masters of Sustaining Innovation

So who are these madmen investing in, and working at, companies like HOT who think they can compete with an R&D behemoth like SAFE.   How could a new company possibly compete?   What are these people thinking?

The answer lies in considering what big companies spend on, and the kinds of innovation at which they excel or struggle.   Clay Christensen did a phenomenal job of framing the issue in his research.   Big companies are very well set up to serve their existing customers and established markets.   They generally pursue “sustaining” technologies that introduce new features or improve performance for established products or product categories.   Companies like SAFE work very hard, and generally effectively, to serve their existing customers and their known requirements.   They invest aggressively to retain their position in their established markets.

Indeed, referring back to the CIO Insight article:

“(R&D) investing in 2006 was mostly aimed at shoring up existing product lines and was concentrated among the biggest firms, like Microsoft, Intel and Google.   The biggest technology companies have been pouring resources into product development, but don't look for a burst of innovative new business applications anytime soon.  … To the extent that companies are developing business applications, their focus is more on enhancing existing products than introducing new ones.”

“Adobe is a good example. The software maker typically spends 90% of its R&D budget enhancing products like Acrobat, Photoshop and Illustrator that have been around for a decade or more.   ‘Some of our products are Version 12 or whatever,’ says Leslie Bixel, director of technology programs at Adobe, which spent $540 million on R&D last year. ‘It really requires that we challenge ourselves to innovate and make sure we're close to the customer to deliver new value.’”

So, SAFE spends a lot on innovation.   But you can bet that most of the R&D budget, and the vast majority of the thought cycles of management, go to thinking about:  how to serve requirements articulated around existing products; how to protect the existing business; and where to find BIG opportunities that can meaningfully impact growth.    Against this backdrop, it must be hard to carve off meaningful attention, from the company’s best developers and marketeers, to thinking about a small new market that may or may not emerge.

Large companies have another challenge.   It’s harder for them to do “multi-functional innovation,” where customer-facing functions (like sales and marketing) and product development must work closely together to craft a new product that fits tightly with a new market requirement.   In a large company, sales, marketing and engineering are usually not on the same floor – and maybe not in the same building.   Product planning is not all that fast, and the communication from customer to product requirements is “lossy” – like that game of telephone we played as kids.   It can take months or quarters for a new product to win budget approval.   And “I think this is something a few customers will need in a year or two” doesn’t usually compete favorably for that budget for the multi-million dollar product line that must be progressed to protect the existing business.

Finally, large companies have to worry about how the new innovation fits with their legacy product and business practices.  They cannot cannibalize their existing product lines.   The products need to be markted and sold by existing channels.   And even if the product is killer and fits a small niche market, it is hard for SAFE to put its best sales/marketing/service muscle behind the new offering.   After all, the field team needs to make quota – so they will always focus on what lots of customers want now – not what a few customers may want next year.

HOT: Masters of Disruptive Innovation

We can now talk about the advantages small upstart companies like HOT have.   Small companies excel at what Christenson nicely describes as “disruptive innovations”.  These innovations create new market opportunities – opportunities that aren’t necessarily large and obvious enough to capture attention of SAFE during the market formation stage when winning, first-to-market technologies are minted.   They are created by discontinuities – big changes in technology or business requirements – that create new customer problems.   The advent of multicore computing, with the multicore software challenge it brings about, is an example of such a discontinuity.

Here, the small company can dominate.   First, the small company has a clean slate.   No large, existing revenue base to distract attention from the new opportunity.   100% of HOT’s attention is focused on the single problem – and a market that is not yet big enough for SAFE can be plenty big enough for HOT.   HOT has a clean slate.   No legacy products with which to integrate.   No installed base that demands ongoing improvements.    And HOT is agile – able to respond quickly to changes in the market and product requirements (because new markets change more quickly than established markets).

And I believe the biggest advantage of a small companies are in “multi-functional” innovations.   New market requirements are best served when a company can build tight communications loops between the customer and product development.  Customer-facing functions and product development in a well-run small company are exceptionally tight.  At Cilk Arts, as an example, we conducted over 40 detailed customer briefings – for a product category that was not yet defined – before we wrote the first line of Cilk++ (to be fair, we should probably count the 15 years of research on Cilk at MIT – but lets ignore that).   The CEO and CTO were in every meeting.

And now that we’re in full-on product development, the loops stay tight.   If our VP Marketing needs to communicate a requirement to our VP Engineering, its 10 paces to his office.   We discuss customer feedback and product requirements in every Monday’s staff meeting, and at our all-hands meetings on Friday.   Point of fact, I believe that superior EMPATHY (defn:  Identification with and understanding of another's situation, feelings, and motives) is a key corporate advantage.   Important enough that it’s reflected in our corporate values.  Empathy in business means listening actively and carefully to customers.  Superior empathy, combined with agility and unity of focus, enables small companies to meet emerging new market requirements more quickly, accurately, and completely than established companies.  

What do YOU think?

• As a technology/product developer, do your experiences match Christensen’s model of which companies are best – and worst –  suited for sustaining (or disruptive) innovation?
• As a buyer or adopter of technology, when have you bet on HOT, and when have you bet on SAFE?