Current immigration policy already stacks the deck against small, innovative companies who don’t have the resources and hiring flexibility of corporate behemoths or offshore outsourcers. Today’s paper-choked H1-B system only accepts visa applications once a year in April, and, if you are lucky enough to win one, finally lets you hire the candidate six months later. (If you aren’t, tough luck. Wait a year to restart the process, pushing out the earliest start date 18 months.) That level of inflexibility and delay is simply not an option for startups, for whom every hire is mission-critical and every day counts. Perhaps as a result, less than two percent of H1-B visas currently go to employees of young venture capital-backed technology companies.

Unfortunately, the Senate’s current reform proposals would create new hurdles for startups without addressing the fundamental process flaws in the current system. Although increasing the number of H1-Bs is a step in the right direction, the new accompanying red tape all but ensures that startups will be squeezed out entirely. Startups have to be lean and nimble to survive. Forcing them to prove they’ve exhausted domestic options before making a foreign hire, or limiting their subsequent ability to make staffing changes, will further hamper their ability to compete. So will requiring them to meet minimum salary levels set for large corporations, since startup employees often happily accept lower salaries in return for equity upside.

Regulations embedded in current reform proposals punish startups for the sins of the outsourcer. Our young technology companies are hiring as many Americans as they can, not pushing valuable jobs offshore or replacing U.S citizens with lower-paid non-citizens. According to a Silicon Valley Bank survey earlier this year, nine out of ten employers in consumer internet and enterprise software said it was a challenge to find qualified workers to fill open positions. Similarly, a recent analysis by the Brookings Institution found 46 percent of job openings that go unfilled for one month or longer require science or technology expertise. Moreover, growth in the tech sector means hundreds of thousands of new, high-paying jobs for Americans. A recent study by the Bay Area Council Economic Institute finds that for every new high-tech job, 4.3 others are created in the local economy.

Rather than reinforcing a broken system, here are some concrete proposals that will level the playing field and turbocharge American innovation:

First, exempt startups (defined as companies less than five years old with fewer than 500 employees) from the H1-B cap, or at a minimum allocate a separate quota to allow them to compete fairly for talent.

Second, allow startups to apply for and use H1-Bs throughout the year to reflect their need for agility.

Third, eliminate requirements for “recruitment” or “non-displacement” attestations or minimum pay levels for startups.

Fourth, create a new “Startup Visa” that allows foreign-born-born entrepreneurs who raise at least, say, $250,000 of funding from qualified U.S. investors to live and start their business here.

And finally, simplify and bring the entire H1-B process online for everyone. (I’m sure there are plenty of startups that would jump at the chance to help with that last objective.)

As the Gang of Eight’s immigration package winds its way through committee, let’s not miss a once-in-a-lifetime opportunity to accelerate American innovation, by funneling the world’s best and brightest to our country’s best and brightest young companies.

The views expressed in this article are those of the author and do not necessarily represent the views of, and should not be attributed to, Andreessen Horowitz.

Sam Flynn: Oh, man, still rocking the pager? Good for you.

- TRON: Legacy

Here’s my card. It’s got my cell number, my pager number, my home number and my other pager number. I never take vacations, I never get sick. And I don’t celebrate any major holidays. 

- Dwight Schrute in NBC’s “The Office”

As Software Eats the World and more and more of our daily activities move online, we depend ever more on IT infrastructure. In a day spent emailing, tweeting, catching up on the news, checking Facebook, shopping on Amazon, watching a movie on Netflix, banking at an ATM — it’s all too easy to forget that underlying all of these “mission-critical” activities are servers, routers, load balancers, switches, storage and millions of line of code.

In a world where we’ve come to view the Internet as a utility, a major website outage is almost as serious as a power outage — and usually affects far more customers. Amazon Web Services had several major outages in 2012, taking down Netflix, Reddit, Heroku and many other sites in July and December. In October, it was the turn of YouTube, Dropbox, Tumblr and Google AppEngine. GoDaddy’s September outage affected up to 5 million hosted websites and 50 million domain names for six hours.

In addition to negative publicity and customer dissatisfaction, downtime now has an enormous financial cost. A 2010 study reported that U.S. businesses suffer an average of 10 hours of downtime per year, at a cost of $26.5 billion. Another analysis suggests that one hour of downtime costs the average business $300,000. If there had been a major outage on our most recent Black Friday, it would have jeopardized $1 billion in online sales.

Dealing with downtime

Of course, modern IT infrastructure has been built for redundancy and is extensively instrumented. Automated tools such as Nagios, Keynote, New Relic, Pingdom, SolarWinds and Splunk monitor every element of the stack and alert engineers immediately to urgent or emerging issues. In fact, today’s machines are very good at detecting and reporting incidents. It’s when those incidents get handed off to humans for remediation that things sometimes break down — because the humans are still using processes and technology that haven’t changed much in ten to fifteen years.

When I was at Loudcloud back in 2001, everyone carried a pager. A small team in our 24/7 Network Operations Center (NOC) would watch for critical monitoring system alerts on big screens and then page the administrator on duty, no matter what time of the day or night. If the administrator couldn’t resolve the issue, they would escalate to developers, who also wore pagers. The process was labor-intensive and error-prone, involving emails, phone-calls, written duty rosters and escalation schedules.

While most other aspects of IT have changed dramatically, incident management in many IT organizations looks remarkably like it did back in 2001. The cloud has done away with the need for many NOCs, and the move to DevOps may mean developers are more directly involved in issue resolution, but the processes are frequently still manual, cumbersome and inefficient. Moreover, today’s large complex systems are never the responsibility of just one person — database administrators, developers, and system administrators all have a role to play — and the more people involved, the more complex and error-prone the process becomes. Reporting of incidents and handoffs from person to person are often done manually via email or SMS. Escalations and problem descriptions are handled via person-to-person phone calls. Engineers consult spreadsheets to see who’s on duty at a particular time. I’m aware of at least one major cloud service provider whose ops people still wear pagers.

PagerDuty

Having studied software engineering at the University of Waterloo and then built and supported large-scale systems at Amazon.com, Alex Solomon, Andrew Miklas and Baskar Puvanathasan set out to bring IT incident management into the twenty-first century. The result is PagerDuty, a modern SaaS-based platform for incident tracking, alerting, and on-call management.

In a nutshell, PagerDuty collects alerts from a customer’s existing IT monitoring tools and alerts the on-duty engineer if there’s a problem. PagerDuty doesn’t replace any particular monitoring tool. Instead, the system sits on top of existing monitoring systems and aggregates all of the errors generated by these tools in a single place.

PagerDuty allows each engineer to configure his or her own customized notification chain. Engineers can opt to receive incident alerts using any combination of phone calls, SMSes, emails and iOS push notifications. So, for example, you could opt to get a push notification immediately when an incident occurs, then an SMS 2 minutes later, then a phone call 5 minutes after that. PagerDuty also allows the on-call engineer to acknowledge, escalate or resolve a triggered incident directly from his or her mobile phone. The company utilizes multiple redundant data centers and SMS and telephony gateways to guarantee reliable message delivery across more than 100 countries.

Incidents in PagerDuty are routed according to an escalation policy. A policy specifies how incidents should be escalated within each team. For instance, you can configure a sysadmin policy to route incidents to a primary on-call engineer and automatically escalate the incident to a secondary on-call if the primary doesn’t answer within 20 minutes. Escalations are crucial to incident response because they add redundancy and ensure nothing falls through the cracks.

PagerDuty lets you build different on-call schedules for each specialization within the organization. For example, you can create one schedule for your database administrators, and another for your network engineers. Incidents can be easily configured to alert the appropriate on-call specialist, ensuring that problems are always automatically dispatched to those who are on-duty and best able to handle them. No more spreadsheets!

Getting customer feedback on PagerDuty proved to be very easy, as it turned out that a large majority of our portfolio companies were using the product — and they were overwhelmingly positive about how it has dramatically simplified and improved their IT operations management. In fact, the company already has several thousand paying customers, including web giants such as Microsoft, Electronic Arts, Adobe, Rackspace and Intuit as well as a growing number of enterprise IT organizations. Overall, PagerDuty has achieved a remarkable amount on about $2 million dollars in initial funding, including generating a substantial and rapidly growing amount of recurring revenue. With a market of almost 10 million infrastructure and application specialists worldwide and multiple ways to expand within the multi-billion dollar IT Service Management segment, this company has a lot of potential.

In closing

The world’s inexorable transition to cloud computing and modern large-scale mission-critical IT systems is creating the opportunity for an exciting new generation of software companies like PagerDuty to play a critical role in its enablement. Many Andreessen Horowitz portfolio companies, for example GitHub, MixPanel, GoodData, CipherCloud and Snaplogic, are members of this class.

As veterans of IT systems management and automation ourselves, we are excited to lead a $10.7 million investment round for PagerDuty and welcome them to the a16z family.

As software continues to eat the world, we spend an ever-increasing portion of our time online. Worldwide, we pass over 35 billion hours a month in the digital world, with US Internet users spending an average of 32 hours online monthly. As the Web has evolved, more and more of that online time is spent in specialized venues such as Facebook, Instagram, Twitter, Pinterest, LinkedIn and Foursquare. While these are fantastic applications, they have a downside, in that they largely exist as parallel, unconnected containers for our personal data. Trapped in their respective silos, our posts, photos, tweets, pins and checkins are largely inaccessible to us from outside. Moreover, creating useful connections between one application and another is far beyond the average user. Sure, most of the most popular web applications now have APIs, but they’re written for the benefit of developers, not people. (Take a look at Instagram’s API documentation, for example.)

Announcing IFTTT

Andreessen Horowitz’s latest investment, IFTTT, (for If This Then That) is out to change all that. IFTTT (rhymes with “gift”) is a simple yet powerful way to create connections between any two web applications, triggering an action on one every time any event you specify happens on another. For example, when I post on App.net, my post instantly appears on Twitter too, thanks to IFTTT. Every time I post a photo (or am tagged in one) on Facebook, IFTTT downloads it to my Dropbox without my even having to think about it. Here’s what that recipe looks like:

Looking for a short-notice ski rental property in Tahoe used to mean checking Craigslist several times a day – this week I just had a simple IFTTT recipe call my cellphone the minute the one I wanted showed up. I no longer check for new movies on Netflix – IFTTT does it on my behalf. While I go about my digital life, IFTTT is in the background, quietly watching out for me.

Like all of a16z’s investments, this one starts with a compelling founder. Linden Tibbets, IFTTT’s co-founder and CEO, started working on IFTTT from his San Francisco apartment in 2010, after three years at design firm IDEO. Struck by how instinctively we know how to use physical objects in creative ways, he set out to enable us to be just as creative with the applications we use in the digital world. To quote Linden, “Much like in the physical world when a 12 year old wants a light-saber, cuts the handle off an old broom and shoves a bike grip on the other end, you can take two things in the digital world and combine them in ways the original creators never imagined.”

“Digital duct tape”

What resulted is IFTTT— described by a recent interviewer as “an idea so alarmingly simple and amazingly powerful… it makes you wonder why nobody thought of it before.” True to Linden’s design roots, IFTTT is visually appealing, approachable and easy to use. In Linden’s words, “IFTTT isn’t a programming language or app building tool, but rather a much simpler solution. Digital duct tape if you will, allowing you to connect any two services together. You can leave the hard work of creating the individual tools to the engineers and designers.”

IFTTT allows people to create “recipes” that connect “channels” (e.g., Instagram, Dropbox, Twitter and 56 other apps) so that “ingredients” on one (e.g., an item with the description “dog painting” appears on Etsy) become a “trigger” for an “action” on the other (e.g. “Add a new line to my Dog Paintings spreadsheet on Google Docs ”).  A glance at some of IFTTT’s channels:

With virtually no promotion, IFTTT has nevertheless achieved remarkable traction since its beta launch two years ago. Its mission statement is to “enable everyone to take creative control over the flow of information.”  People have created over 2 million individual “recipes” to connect their favorite websites and apps in ways that meet their own unique needs. There are tens of thousands of shared recipes for common use cases. Three million recipes are executed every day, addressing individual interests as varied as “Text me if Apple stock drops below $500” and “Post my App.net posts tagged #a16z to Yammer.”

Looking ahead

Although Linden and his tiny team of seven have achieved an amazing amount already, the best is yet to come. With Andreessen Horowitz’s investment, 2013 will bring more and simpler recipes and exciting mobile apps. A new developer platform will enable application developers to create services that connect their application to others in new and powerful ways, opening up new functionality to users with virtually no diversion of internal engineering resources. Perhaps most exciting of all is the role IFTTT can play in the emerging Internet of Things. As everyday objects from fridges to shoes to weighing scales become equipped with communicating smart sensors, individuals’ need to create useful connections and information flows between them will far outstrip their developers’ capacity to build them. I don’t know exactly when my fridge will be capable of knowing I’m running low on milk and contacting Safeway to order more, but there’s a pretty good chance an IFTTT recipe will be involved when it happens.

Time to step up:  Hiring the VP of Sales

For the technology founder-led companies we love to back at Andreessen Horowitz, recognizing the need to “hire different” is particularly important when the company is transitioning from initial product development mode into sales mode.  If you’re an a16z founder, you’re very likely an engineer or computer scientist.  Virtually all of your early hires will have been engineers or developers.  While you’ve hopefully hired people who are smarter than you, they may not be that different from you.  Now you’ve got a product, it’s time to hire a VP of Sales.  Prepare to hire different—very different!

A typical sales culture is different from a typical engineering culture in almost every aspect—from personality types to values to dress code to working hours.   Introducing sales DNA is highly likely to clash with your engineering culture.  That prospect will feel quite uncomfortable, but is absolutely critical if you’re serious about building a real company.

While on an intellectual level you may recognize the need to hire an executive who’s different, it can be really hard to act on it in practice.  Most of us are naturally more comfortable with people who are like us.  If as a first-time tech founder CEO you feel entirely comfortable with that VP of Sales candidate, she’s probably not the right one.  The right candidate will feel risky to you from the cultural fit point of view.  So, you’ll want to know what the team thinks.  However, the interview feedback from the team is likely to make you even more uncomfortable.  Here are some real examples of technical team feedback from interviews of highly qualified sales candidates:

“I’m worried that he’s rough around the edges and would clash with our people.”

“His tone was very A-type salesman.”

“I fear for the culture if she were to come aboard.”

“He had very limited understanding of our product.”

“I don’t think he has much idea of what we do and how we do it.”

“I think he could generate massive revenue, but he’d destroy the culture by building a Salesforce-type sales organization.”

So, hiring the right sales leader will involve not just overruling your own emotions, but rejecting the strongly held opinions of several members of your team.  Remember you’re not recruiting her for her knowledge of the company, but for her knowledge of the outside world.  As my partner Ben puts it: “Generally, you want product leaders with superior internal knowledge (knowledge of the code base, knowledge of the culture, knowledge of the people).  With sales people, it’s the opposite—they need to have external knowledge (knowledge of customers, purchasing processes, customer org structures, customer cultures, . . .)”.

Two-way street

While you and the team are trying to evaluate this unfamiliar animal and assess whether you can handle it in your habitat, don’t forget she’s evaluating you too.  Just as you’re asking yourselves things like, “Will this person destroy our culture?”, she’s asking herself, “Are these guys serious about building a business, and will I get the support I need to build a winning sales organization?  Will this CEO have my back with the company and the board?”  She won’t need everyone’s buy-in day one—a good VP of sales will earn that over time—but she will need full and sustained support from you and the board to put in place the processes, people and sales culture that can transform a great product organization into a great business.   This is your time to lead.  The way you handle the recruiting process will speak volumes about whether you’re the leader she wants to bet her career on.

Hiring different at Opsware

At Loudcloud/Opsware, we went though three VPs of sales in the first four years of the company.  Each one of them was well liked and an excellent fit with the culture, but not one of them was able to consistently hit our quarterly sales targets.  By 2003, looking for our fourth sales leader in as many years, we met a guy named Mark Cranney[1].  He seemed very well qualified, and he gave us several pages of references to call if we wanted proof.  However, virtually everyone who interviewed him had similar feedback:  “He might be a great sales guy, but there’s no way he’d fit here.  Way too risky.”

Ben Horowitz and I called every one of his references.  They confirmed that Mark was an exceptional sales leader who had consistently achieved great results.  But could we take the risk of introducing such a culturally different executive into Opsware’s strong culture?  Only Ben and I seemed to think so.  We hired him.

Hiring Mark, and the many sales professionals he added to our ranks, was indeed a shock to the system.  It created significant tension in the company at times, as sales culture met engineering culture.  It was a tension that we badly needed—a healthy tension between the demanding outside world of customer needs and competitive pressures and our sheltered inside world of PRDs and predictable development schedules.  Mark was unreasonable—he told it like it was, not how we wanted it to be.  It stretched every part of the company, from engineering to marketing to legal to finance.  Gradually, we became a more customer-driven organization and we started making our numbers.  Over time, cultural discomfort shifted to mutual respect and even to affection.  In the years that followed, we won hundreds of the world’s most demanding enterprise IT accounts, met their needs with market-leading products, outpaced a brutally tough competitor, and grew bookings and revenues at a rapid pace.  For Opsware, hiring different was the key to unlocking the company’s true potential.

The courage to challenge

Strong leaders don’t just hire people who are smarter than them, or different from them.  They also look for people who have the courage to challenge them, and they create a culture that encourages people to do so.  Ironically, being surrounded by lots of smart people makes it harder to speak up, particularly for more junior team members.  “If all of these smart people think X makes sense, who am I to challenge them?”  The more senior you are, the less likely you are to be challenged, and the more you will have to work to encourage it.  “She’s the CEO—she must know what she’s doing.  I’m not going to question it.”  I don’t know about you, but if I were the emperor, I’d want someone to tell me before I walked down the street naked.  Like the emperor, you won’t hear diverse opinions if you don’t actively solicit them—and resist the temptation to slaughter them with your ferocious intellect the minute they’re expressed!

At Andreessen Horowitz, we’re fortunate to have a lot of very smart people, more senior and more junior, around the table when we talk about companies we’ve just met.  It takes real courage to speak up when everyone else seems to love (or hate) some opportunity we’ve just seen, but we particularly want to hear that different view.  We work hard to encourage people to speak up.  The best investments are often controversial, and vigorous debate leads to better investment decisions.

The same applies to business decisions.  As CEO, it’s your job to make the call.  Having people around the table with the background and the courage to be controversial makes it more likely you’ll make the right one.  As your company grows, recognize that you’ll need to hire different, and strive to build a culture that encourages and rewards constructive challenge and diversity of opinion.

I just came back from a short stay in Rio de Janeiro, Brazil.  Like its fellow BRICs China and India, Brazil has experienced a massive economic boom over the past decade.  While the boom has lifted tens of millions of people out of economic misery (40 million have entered the middle class in Brazil alone), the increasing challenge for the world is to sustain such growth without intolerable pressure on the earth’s increasingly scarce resources. While I was in Rio, the city hosted over a hundred heads of state for “Rio+20”, the United Nations Conference on Sustainable Development. While the conference has passed, the immense challenge of sustainable global development remains—and sustainable agriculture is a particularly critical imperative.  As the UN’s Rio+20 website puts it:

…Right now, our soils, freshwater, oceans, forests and biodiversity are being rapidly degraded. Climate change is putting even more pressure on the resources we depend on, increasing risks associated with disasters such as droughts and floods.  Many rural women and men can no longer make ends meet on their land, forcing them to migrate to cities in search of opportunities.”

The Green Revolution – Part I

Forty years ago, the world was faced with the similarly daunting challenge of feeding the exploding populations of India, China and other developing nations.  A 1968 bestseller predicted that “India couldn’t possibly feed two hundred million more people by 1980″ and “hundreds of millions of people will starve to death in spite of any crash programs.”  In fact, technology saved the day, in the form of extraordinary, new high-yield grains and seeds, powerful new fertilizers and modern agricultural management techniques.  Far from succumbing to famine, India and its peers became first self-sufficient and then net exporters of food, as a technology-powered “Green Revolution” enabled a massive increase in agricultural productivity.

The Revolution’s Serious Costs

Although the Green Revolution was a tremendous success, it was not without serious costs that are being felt with ever-increasing intensity today.  One of these costs is the excessive and indiscriminate use of chemical fertilizers around the world, with severe consequences for arability and for air and groundwater.  (For example, nitrogen causes algae blooms that deprive water of oxygen and kill marine life.)  China provides a case study: While farm yields in China have increased 40 percent since 1980, chemical fertilizer use has increased by 225 percent.  Chinese farmers now use over three times as much nitrogen fertilizer per hectare as U.S. farmers, but produce 30 percent lower yields.  Regulatory restrictions are proliferating worldwide.

Meanwhile, the challenge of feeding the world is back on the front burner.  The same UN report says:

A profound change of the global food and agriculture system is needed if we are to nourish today’s 925 million hungry and the additional 2 billion people expected by 2050.

The Green Revolution – Part II

The key to meeting this challenge will be to achieve further dramatic increases in crop yields, but in a sustainable way.  American farmers are leading the way, in a new technology-powered Green Revolution, investing in ultra-modern tractors equipped with smart devices that automatically control seed and fertilizer dispersal.  According to a recent fascinating Wall Street Journal article, “These modern tractor cabs have come to resemble airplane cockpits more than the seats of old-fashioned tractors.  Farmers can plant or fertilize a whole field without touching the steering wheels, and yield data from sensors in combines—the vehicles that harvest crops—help refine the plans for the next season’s planting.”

Introducing Solum

Solum, Andreessen Horowitz’s latest investment, is poised to play a fundamental part in this exciting shift to data-driven precision agriculture.  Founded by three brilliant young men with PhDs in Applied Physics from Stanford and ties to the American Midwest, Solum has invented a radically more accurate technology for testing farm soil, enabling farmers to measure actual nutrient content and apply fertilizer on a targeted and highly granular basis.  Solum’s platform enables farmers to correlate nutrient measurements and fertilizer application to actual yields, in a constantly improving feedback loop.  Over time, the result for farmers should be a “virtuous circle” of increasing crop yields driven by ever-smarter and environmentally sustainable use of fertilizer, water and other precious resources.  In essence, Solum’s technology will provide the data to drive farmers’ new intelligent machines, and the software to manage their application on a large scale.

While the potential societal and environmental benefits are enormous, a powerful profit motive will also drive global adoption of precision agriculture technologies such as Solum’s.  According to an expert in the WSJ article, “Improving how the seeds are planted and the soil fertilized can increase corn yields by several tens of bushels per acre.  At current prices of near $6 a bushel, every 10-bushel-per-acre increase in the yield on a farm of 2,000 acres would translate to $120,000 in additional revenue.”  The potential is truly global.  In its quest for more sustainable agriculture, China’s Ministry of Agriculture plans to test fully 60% of its arable soil over the next five years.  Brazil, which is rapidly becoming the world’s breadbasket, is investing heavily in advanced agricultural technologies and increasingly determined to make its remarkable advances sustainable.

“Software Eats Dirt”

In his 2011 Op Ed piece “How Software is Eating the World”, my partner Marc Andreessen vividly described a dramatic and broad technological and economic shift in which software companies, often based in Silicon Valley, are transforming almost every sector of the global economy.  As economic sectors go, agriculture is enormous—in fact, agriculture is the single largest employer in the world, providing livelihoods for 40 percent of the global population.  In our internal discussions, our affectionate codename for Solum was “Software Eats Dirt”.  We believe this is a company with the potential to revolutionize agricultural production literally from the soil up. I’m excited to be joining Solum’s board and we’re honored to support such an ambitious and important mission.

But screw your courage to the sticking-place, and we’ll not fail.
—William Shakespeare, Macbeth

In my last post, I described how, at Opsware, we would step back from the fray about once a year to go through a strategic review of our situation with the board. In November 2005, our conclusion was to keep on building an independent company, but in early 2007 another review led us to a different conclusion. Our business was strong, but meeting quarterly expectations was difficult and the market wasn’t rewarding us for the growth and leading market position we had achieved. We were at a decision point: Plow even more money into R&D and sales expansion or explore a sale. We decided to explore a sale, but only if we could achieve a top-dollar price for our shareholders.

The dance begins: May 2007

Our February update meetings with the major enterprise IT players had piqued their interest and led to several follow-ups.  The key was to get someone to set an initial price. On May 22, we went back to the Opsware board with an update. A company we’ll call Company 1 had just offered to buy the company for $11/share. That was a 38% premium—good, but not exceptional. Once again, we discussed the opportunities, risks and shareholder value implications of continuing in the business. We debated long and hard: What was our number?

We decided to send a strong message to anyone considering us as a target: We told Company 1 that the Board had rejected its $11/share offer and would not give serious consideration to any new offer below $14/share. For good measure, we took the opportunity to notify the other companies who’d followed up with us (HP and several others) that we’d received an offer but were not interested in discussing any deal below $14. Since that was a 75% premium over our current price, Company 1, HP and everyone else told us they were out. We sent all of them letters requesting the return or destruction of any information we had shared. Those brief dreams of exit seemed to be just that—dreams. We went back to the slog of making yet another quarter.

Then, almost a month later, the CEO of Company 1 came back offering $13.25/share. Yes! We were getting close to our magic number! We reconvened the board for a serious, fact-based discussion of staying the course versus selling in this price range. Our analysis suggested we would have to grow revenues at least 75% annually to exceed $15/share. That seemed impossible—Wall Street consensus was 28%.

In addition to upside opportunities, we discussed the risks of staying standalone. (Note #6, macroeconomic slowdown: We had no idea how real a risk that would turn out to be just months later.)

It was clear: We should exit. We decided to stick to our guns and drive for a deal at $14/share, leveraging competition and urgency to get there. The first step was to get Company 1 to $14, which we did two days later. Then we notified the others, notably HP, that we had an offer at $14. A whirlwind of meetings and calls followed with the various contenders.

While we had discussions with 10 companies, by July 18 it became a two-horse race between Company 1 at $14.05/share and HP at $14.25/share, each acutely aware of the other’s interest. To keep up the pressure, we negotiated detailed agreements with both in parallel right up to the end, with a real prospect of getting to $15 or more. Although exhausted, we were on a massive emotional high—this was going to be a fittingly impressive end to an eight-year odyssey!

Crisis strikes: July 18

Then, with 24 hours to go, we had an inconceivably ugly crisis. In diligence, Company 1’s auditors, Ernst and Young (EY), challenged the way we had accounted for three customer contracts, despite the fact that EY was also our auditor and had ordered this accounting. Unbelievably, two regional offices of EY were disagreeing with each other. Called to arbitrate, the national office sided with Company 1’s regional office and informed us we would have to restate our publicly-filed financials for the past several years! While the issue was entirely technical and the amounts immaterial, the prospect of a restatement was horrifying. Not only had our glorious deal suddenly vaporized, but by chasing it we’d triggered something that might panic investors and cut our stock price in half. In the blink of an eye, our likely outcome had plummeted from $15/share to maybe $4/share. We were devastated.

Company 1 was seriously spooked and asked for a couple of days delay. We knew we’d have to tell HP. We put the best face we could on it—“technical issue, immaterial amounts, no impact on value, etc.”. Hearts in our mouths, we waited for their reaction. To our intense relief, they reacted with concern but didn’t walk. However, our leverage was gone and our deal seemed on the brink of collapse.

Our only hope was to get the three customers to agree to a minor change in their contract language that would enable the original accounting treatment. Theoretically possible, but we were almost out of time, and these were huge corporations. Working through the night in a hot stuffy conference room, we started emailing and calling our executive contacts. Unbelievably, in a testament to the strength of our customer relationships, all three mustered their attorneys and got it done in 24 hours. Restatement avoided! We were back on track—but with only one bidder currently left at the table. Could we still get it done?

Psych test: July 19

Perhaps sensing advantage, HP execs suddenly dropped their offer to $13.75 in a tense 3 pm meeting the next day with Ben and me in their executive offices. We were furious, but we had come this far and we had no other live bidder. Maybe we should just suck it up and accept that EY’s ineptitude had cost us a few hundred million dollars?

Then, thinking on the fly, we realized it wasn’t that simple. How we reacted would be critical, not just to the deal price but to the very survival of the deal itself. Anything less than an unequivocal rejection would signal to them that they were in the driver’s seat. If we were now willing to accept $0.50 less per share, then why not $1 or $2 or $3 less? If there was really no one else at the table, why not drag out the process and wear us down? After all, this was a very expensive deal—why not terminate and restart at a much lower valuation?

We knew HP was the only one currently at the table, but they couldn’t be sure of that. We knew it was time to “screw our courage to the sticking-place”. We informed them that we had scheduled a final board meeting for 6:30 pm and we would not be recommending the deal with HP for anything less than the original price.

With Company 1 still AWOL and no word from HP, we started the board meeting. What the hell were we going to tell the board? Had we killed our deal over 50 cents a share?

We needn’t have worried. By 6:35 pm, my cellphone was lighting up with calls from HP’s M&A guy, worried they were about to lose the deal. We made him wait. When I finally called him back about 7 pm, he told me they were willing to offer $14.05. Ha! Although some board members and fellow execs strongly urged us to take it, we resolved to go back for the final 20 cents as a matter of principle and to preserve the crucial psychology of a hard-won deal. Fifteen minutes later, we had an agreement to sell Opsware to HP for $14.25/share in cash. We were drained, but elated. Our eight-year journey was at an end.

In summary

Although we never built Opsware with the intent of selling it, acquisition ultimately turned out to be the best outcome for our shareholders and our employees. While the company was clearly an attractive target, the significant premium we achieved in the sale was also the result of the company’s ongoing investment in strategic business development and a tightly executed deal process. And in knowing where the exits were.

So you got to let me know
Should I stay or should I go?
—The Clash

To lead you to an overwhelming question….
Oh, do not ask, “What is it?”
Let us go and make our visit.
—T.S. Eliot, The Love Song of J. Alfred Prufrock

Taking the exit

On July 23, 2007, HP announced it was buying Opsware for $1.6 billion in cash. By any measure, it was a very attractive deal for Opsware shareholders. The acquisition price of $14.25 a share represented a 74% premium over the prior six-month average, and a forty-fold return for anyone who had bet on us at our low point in October 2002, when only an intensive investor relations effort had saved us from NASDAQ delisting. Almost five years later, the acquisition multiple of almost 16 times trailing revenue still far exceeds that of any other billion dollar-plus enterprise software acquisition—ever.

The hard work of hundreds of employees contributed to making us the clear high-growth market leader in what was finally an important enterprise software category, but the significant premium we achieved was also the result of a multi-year, strategic business development effort designed to turn a merely good outcome into a truly exceptional one. That business development effort was built around five key principles that you, too, should consider as you build your company:

1. Always know where the exits are: Take the time to build relationships with potential acquirers. You never know when you may need them.
2. Step back from the fray occasionally: Review the company’s strategic situation with the board every 12-18 months and evaluate the alternatives with quantitative and qualitative rigor.
3. In evaluating your alternatives, ask yourself how you would feel if the environment were to change radically: Days before we signed the HP deal, the Dow hit 14,087—its highest level ever. Within months, hedge funds and mortgage companies started imploding and we entered the worst recession since the great depression. Just three months later, a sale at even $8/share would not likely have been achievable.
4. If you do decide to sell, a meticulously executed competitive process is key to a successful M&A outcome: This SEC proxy statement description of Opsware’s process details discussions with 10 companies, stretching over nine pages. Every one of those contacts was tightly scripted and carefully orchestrated to drive to the ultimate successful outcome.
5. Creating and managing the acquirer’s psychology is critical: The key is subtly to convince the acquirer that they have no option but to acquire you, but that you have multiple attractive alternatives—and to reinforce that impression in every interaction, no matter how small, until the deal is definitively done.

“Your best exit may be behind you”

As I’ve discussed in previous posts, the best CEOs leverage strategic BD to cover critical flanks not well covered by other functions—helping to navigate crises, driving acquisitions and strategic partnerships, leading international expansion. No great entrepreneur sets out to build a company to be acquired, but as the company grows, it’s important to have options and to understand them well. Perhaps the most important role of a strategic BD exec is proactively to make sure the company always has options and that the CEO and board always know where the exits are. They tell you on an airplane that “your best exit may be behind you”, but that’s not something you ever want to hear in business.

Stepping back from the fray: November 2005

At Opsware, 90% of our board discussions were naturally about the day-by-day work of building a great company: quarterly bookings, wins and losses, product plans and challenges, financials and so on. However, about once a year we would make a point of stepping back from the daily battle for a more strategic review of our situation. As part of this review, we would address more fundamental questions.  For example:

• How do we feel about our market?
• How is our competitive position?
• Who are potential acquirers of the company, and what’s their current state of mind?
• What are the major opportunities and risks facing us?
• Bottom line: What’s the likely value of staying the course versus exiting?

Prior to these board discussions, we would make a round of senior visits to HP, BMC, Oracle, EMC and other behemoths to give them an update on the business.  While our ostensible purpose was to explore the potential for a partnership, our real objective was to understand (and pique) their interest in our space and to make sure we’d have an open door if we should ever decide to explore selling.

I’ve linked our board presentation from November 2005 here. As you can see, it’s a pretty thorough analysis. Here’s the executive summary:

In other words, let’s keep marching!

Stepping back from the fray: February 2007

In early 2007, Marc Andreessen, Ben Horowitz and I made another set of visits to the usual suspects in preparation for a new strategic review with the board. We had made major progress since we had seen them last: exceeded $100M in annual revenue, fired up a productive distribution deal with Cisco, added storage automation, expanded internationally. As before, we went through a pretty slide deck that showed a winning company in a strategic category, with a final slide that mused vaguely about partnership. Left hanging in the air at the end of some of the meetings, we could sense the “overwhelming question” in the mind of the big company CEO: “Should we buy these guys before someone else does?” Several of them requested follow-up meetings.

For us, too, the possibility of actually selling the company was of more than theoretical interest this time around. Meeting Wall Street’s expectations was still a challenge every quarter. Our one serious competitor, BladeLogic, was about to go public. Despite five years of strong execution, our own stock price seemed stuck in a narrow band around $8/share due to continued heavy investment in R&D and sales expansion. Strategically, we were now at a critical juncture: Invest even more heavily in new capabilities like monitoring and enhanced support for virtualization—or capitalize on our current position and the strong interest from acquirers. That exit looked pretty tempting.

That said, selling for a typical enterprise M&A premium of 25-30% didn’t feel compelling to us or the board. We hadn’t escaped the jaws of death and worked this hard, for this long, to build a great company, only to sell it for 10 bucks a share. We were resolved: The only way we’d consider selling was for a price much higher than what we felt we could achieve in the standalone case.

Here’s the recommendation from our February 2007 board update:

That led us into a dramatic three-month dance with a series of suitors that would test our negotiating and business skills, as well as our resolve.

Next up: The dance begins

The reason is simple.  We are fortunate to work with some of the best entrepreneurs and technologists in the world, and in the process help create great and valuable companies.  That activity, done well over decades, can generate a lot of money that can then be productively deployed philanthropically back into the society that makes it all possible.  We love participating in this process, and we hope that our philanthropy can, over time, help make the world a better place.

As an initial catalyst, we are making an immediate group donation of $1 million to a set of six vital Silicon Valley-related nonprofit organizations.  Those causes, and their respective sponsors, are:

Ben and Felicia Horowitz: Via Services
Jeff and Karen Jordan: Ecumenical Hunger Program
John O’Farrell and Gloria Principe: Second Harvest Food Bank
Marc and Laura Andreessen: Fresh Lifelines for Youth
Peter and Martha Levine: Canopy
Scott and Pamela Weiss: The Shelter Network

Signed,

Ben, Jeff, John, Marc, Peter, and Scott

SUNNYVALE, Calif., Feb. 13, 2006
Opsware Announces Worldwide Distribution Agreement with Cisco

These two headlines sound pretty similar—“Small company partners with giant company to reach a bigger market”—but they led to two very different outcomes.  Our 2003 deal with HP didn’t generate a single dollar in revenue, whereas our 2006 agreement with Cisco drove tens of millions of dollars in sales and helped to make Opsware the uncatchable leader in data center software.  Why did one succeed spectacularly while the other never took off?

As a startup with the best product, your challenge is often getting it in front of enough customers and getting them to buy.  In theory, striking a deal to have an HP or an EMC or a Vodafone sell your product to their customers is the way to cover the market and exponentially increase sales velocity.  In practice, however, most “David-Goliath” distribution deals turn out like our 2003 HP deal: great PR, but not much else.  Here’s the way it typically plays out:

• The deal is announced with great fanfare and high internal and external expectations.
• Goliath needs product changes, training and lots of help to even attempt to sell your product.
• Goliath deluges your already overloaded people with feature and support requests.
• You can’t justify assigning dedicated people to support Goliath because you can’t bank on any new revenue.
• Goliath sells nothing, or even worse, ends up competing against your sales team for customers you would have won directly at a higher margin.
• The “partnership” quietly withers away, leaving a damaged relationship with Goliath and a bad taste inside the company.

The temptation for some startups is try to “make it up in volume”—sign as many distribution partners as possible on the basis that none of them is likely to deliver much on their own.  While that degree of over-coverage might feel temporarily reassuring, it only multiplies the challenges exponentially in practice.  The only thing worse than having no partners is trying to manage multiple ineffective partners competing with each other in the market and drowning the company with their demands.

In my experience, one well-constructed, high-impact partnership is better than a hundred run-of-the-mill arrangements.  Distribution partnerships take an enormous effort to make work.  It’s just not realistic to have many of them, which is why picking the right one and constructing it intelligently is fundamental.  At Opsware, we only had one significant distribution relationship over eight years, but it had a massively positive impact on the company.

Time to find a partner

It was early 2005, and things were starting to go pretty well for three-year-old Opsware.  We had just closed our $33 million acquisition of Rendition Networks, giving us a lightweight but powerful network automation software product that we christened Opsware NAS (Network Automation System). NAS would allow us to penetrate and rapidly deliver value to new accounts, then upsell them to our considerably heavier and more expensive Server Automation System.  (Our average NAS deal size was $115k, took a month to sell and a couple of weeks to deploy, whereas SAS deals averaged $765k and could take a year to sell and six to twelve months to deploy.)

It made sense to market and sell NAS as aggressively as possible—but we faced a challenge:  Our direct sales force was small, only covered US and UK, and had few relationships with network buyers.  We needed a big partner with the buyer relationships and global coverage we lacked.

The right partner

The right partner was obvious:  Cisco.  So was the product: NAS.

Thanks to our investment in strategic BD (see Part I and Part II of this series), we’d been calling on Cisco since the founding of Opsware, from the C-suite to mid-level data center managers.  Now we finally had something to talk about.  For all its pre-eminence in router and switch hardware, we knew from customers that Cisco’s management software capability was weak for its own devices and non-existent for, say, Juniper devices in the customer’s network.  We had the market-leading, cross-platform product.  Within days of closing the acquisition, we visited the Cisco SVP responsible for management software and offered him the solution his customers were demanding.

Mission clarity

Before you can construct a high-impact deal, you have to be crystal-clear on what you want to achieve.  We spent a substantial amount of time up front to achieve internal clarity and consensus across the company on what we wanted from a partnership.  These were our main objectives:

• A real commitment by Cisco to market and sell a lot of NAS product, at an attractive margin to us.
• Opsware’s right to target every Cisco NAS customer for an Opsware SAS sale.
• Opsware branding on the Cisco NAS product.
• Minimal product changes and a single code base.
• Enough assurance of financial potential to allow us to invest in the support Cisco would need to succeed.

In a seemingly endless series of meetings in Cisco’s blue-gray conference rooms over the spring and summer, we gained a good understanding of Cisco’s main objectives.  They wanted:

• The ability to sell NAS to any Cisco account, with active support, not competition, from Opsware.
• Cisco-branded product.
• Freedom to sell at any price they chose, including throwing in the NAS software free as part of a big equipment sale.
• “Insurance” against a surprise acquisition of Opsware by a competitor.
• Access to Opsware NAS source code, and even the ability to make derivative works.   (As you might imagine, this one caused us a lot of heartburn.)

All through the fall and early winter, we shuttled back and forth between Tasman Drive and Opsware’s office on Mathilda Avenue, alternately negotiating a deal structure with the Cisco software team and updating and strategizing with our sales and product organizations to make sure we were staying true to our objectives.  Some issues, like source code, were extremely complex.  Nonetheless, like a figure emerging from the fog, a compelling deal gradually took shape.

On February 13, 2006, a year after closing the Rendition acquisition, we announced a worldwide distribution agreement with Cisco.  Cisco’s enterprise sales force would sell our NAS product, rechristened Cisco Network Compliance Manager, to their worldwide customer base.  Opsware’s sales force would follow their Cisco counterparts, supporting their efforts to sell the network product and in the process building the relationships and laying the groundwork for a subsequent million-dollar-plus server software sale.

Behind the headlines – a deal with teeth

While the press and public endorsement by Cisco were valuable, what made this partnership work unusually well in practice was a smart deal structure designed to achieve both sides’ objectives:

Minimum revenue commitment:  It took us almost nine months, but we persuaded Cisco to make a binding three-year quarterly revenue guarantee that totaled almost as much as we had paid to acquire Rendition.  We helped our Cisco counterparts to understand that this would be a win-win:  Guaranteeing several million dollars a quarter to Opsware would give Cisco huge motivation to drive sales.  In turn, it would give us the assurance we needed to back off our own NAS sales efforts and do everything—including assigning some dedicated sales and product headcount—to make our partner successful.

Substantial revenue share: We secured a revenue share and pricing floors that made a Cisco NAS sale almost as financially attractive to us as if we had sold it directly—because Cisco could actually charge more for the software than we could.

Co-branding:  The product was marketed as Cisco Network Compliance Manager, “built on Opsware automation technology”.

Channel-neutral compensation:  To motivate Opsware’s sales people to support their Cisco sales counterparts instead of competing with them, we compensated them for every Cisco sale in their territory.  While this cost us real money, it made sense because every Cisco sale opened the door for us to sell a much bigger, higher-margin product.

Acquisition insurance:  We committed to give Cisco advance notification of any agreement to sell the company, and the opportunity to enter the bidding if they wanted to.  (Note this is very different from a Right of First Refusal.)

Source code license:  We gave Cisco a three-year license to view our source code and build on it, but came up with an innovative licensing approach that protected us fully and even generated millions in additional revenue.  In any case, we bet they would never actually do anything with the code, and we were right.

Goliath delivers…

Within nine months, Cisco was selling large network software deals into accounts like DHL, Costco, Sprint, USPS and Telstra, and a short time later, they were selling enough to exceed the quarterly minimum revenue guarantee to Opsware.  The Cisco deal would go on to generate over 25% of our bookings.  Cisco’s massive distribution power vaporized our NAS competitors.  Just as important, every NAS sale opened the door for our SAS product in accounts we had never penetrated before, providing a major thrust to our high-margin server business.

Seen in a Cisco corridor

…with a little help from his friend.

Of course, there were many challenges along the way.  Cisco people needed lots of training and support to sell the product.  There were occasional conflicts over accounts that had to be arbitrated.  We ran a separate weekly pipeline call with Cisco to track and drive their sales and direct our follow-on server sales calls.  Product change and bug fix requests from Cisco had to be triaged and folded into our roadmap.

However, we were able to handle these demands with equanimity, because we had a couple of dedicated heads in sales and product management to handle them, and a large guaranteed revenue stream that helped the whole company to realize this was a deal worth supporting.

In summary

Like Opsware’s 2003 HP deal, the vast majority of David-Goliath type distribution deals don’t deliver beyond PR^[1].

On the other hand, like Opsware’s Cisco deal, the right distribution deal with the right partner can be truly transformative.  In formulating your partner strategy, consider four golden rules:

1. Quality trumps quantity:  One well-constructed partnership will have far more impact, and be far more supportable, than lots of toothless arrangements.
2. Mission clarity:  Invest the time up front to get crystal-clear on your objectives and those of your partner.
3. Demand hard commitments:  Big companies are easily distracted.  Binding commitments focus the mind and create incentive to deliver long after the deal is signed.  (Note this means finding a senior executive sponsor with the authority to commit.)
4. Make hard commitments in return:  You will need to invest substantially in making your partner successful.  Commitments like sales and product support and channel-neutral compensation are worth making in return for the right commitment from your partner.

Only a few startups invest in the ongoing BD effort to systematically cover the landscape of potential partners and create that game-changing partnership when the opportunity presents itself.  For Opsware, that investment paid off three years after our start as a software company.  Like the Loudcloud-EDS deal and the M&A campaign that followed, the Cisco partnership provides another example of the power of strategic business development. 

Opsware V1: Single product, single customer

The EDS transaction closed in mid-August, and we started into the fourth quarter of 2002 in much better shape than we’d entered the first quarter.  That said, we still had plenty of challenges.  We were a single-customer, single-product company.  Moreover, the product—software that automated the administration of servers—was powerful but far from ready for broad commercial adoption.  It had been developed solely for internal use by skilled (and somewhat forgiving) Loudcloud system administrators in a controlled environment.  It was difficult to sell and hard to implement, requiring significant process change on the part of the customer’s IT staff.  It wasn’t easily consumable in small chunks, leading to protracted sales cycles and long implementation periods.

Of course, we had 60 great engineers and plenty of cash, and they lost no time starting the work needed to make the software easier for our small but talented sales team to sell.  While they did that, the BD team led the development of what would become a selective but highly productive M&A strategy to turn Opsware into a broad data center automation platform.

Kicking off the M&A strategy

Working closely with the product and sales teams to understand needs and build consensus, we started identifying and prioritizing categories adjacent to our core server automation category and tracking the key players in each.  Together, we developed a set of criteria for acquisitions, including technical factors such as architectural compatibility and APIs as well as non-technical factors like location, team size, culture and financial profile.  As a general theme, we wanted products that fit with our core proposition of data center automation and were compatible with our server product, but were easier to sell and implement, opening the customer’s door for a subsequent server automation sale.  We briefed the board repeatedly to prepare them for specific acquisition proposals to come.  Thus armed for battle, we set forth.

Going into battle

Any good strategy allows for opportunism and, as it turned out, our first acquisition was highly opportunistic and somewhat unusual.  Tangram was a 20-year-old company, publicly-traded on the OTC “pink sheets” exchange, with a market cap of about $10 million.  Their product was IT asset management software, and while their revenues were growing slowly, their 200 enterprise customers were very satisfied with the product.  We seized the moment and picked up $10 million in annual revenue for stock valued at less than $10 million—a bargain.  The deal brought us a complementary, consumable product, 200 new prospects for server automation software and a low-cost base in Cary, North Carolina, that soon became our “offshore” location for support and sustaining engineering.  In short, a good start to our M&A plans, just four months into our new life as a software company.

Our next acquisition, a year later, was much more strategic.  Our customers were beginning to like our server automation product and wanted to know if we could solve their next pain-point: automating the configuration of network devices such as routers and switches.  We identified the four leading players in network configuration management, visited and screened them all, and ran a competitive process to get them competing to sell to us.  (It’s always better to have them trying to convince you to buy than for you to be trying to convince them to sell.)  We got the one we wanted, Rendition Networks, for a combination of cash and stock valued at around $33 million.  In addition to 30 new enterprise customers and a strong engineering team in Redmond, Washington, we would now have an attractive starter product and a combined server-network automation offering that put us further ahead of our only credible single-product competitor, Bladelogic.  Within a year, Opsware Network Automation was generating over 20% of our bookings.  Even better, the Rendition purchase positioned us for a major OEM deal that would further accelerate Opsware’s breakout lead.  I’ll talk about that in the next post.

With servers, network devices and asset management in the bag, storage was the missing piece of the puzzle.  That proved to be harder—there were few storage automation startups to begin with and none that looked like a great fit.  We ultimately did acquire a small startup named Creekpath Systems in 2006 for $10 million in cash, closing their small Colorado HQ and moving a few team members to Redmond.  Unlike our other acquisitions, storage automation proved just as hard to integrate and commercialize as it was to find an acquisition target, and this was the least successful of our purchases.  Notwithstanding that, we now had a unique full data center automation story: servers, network and storage.

Our fourth and final acquisition was in an emerging category called Runbook Automation.  Having automated individual functions such as server provisioning or router configuration, our customers now wanted to automate entire IT workflows across servers, networking and storage, integrating with other systems such as monitoring or ticketing.  Here there were just two viable targets: RealOps in Herndon, Virginia, and iConclude in Bellevue, Washington.  We knew we wanted iConclude, but again a competitive process of negotiating with both companies helped to drive a speedy transaction at a fair price of $54 million in cash and stock in March 2007.  (iConclude’s founder and CEO, Sunny Gupta, now runs Apptio, an Andreessen Horowitz portfolio company).

Although we only made four acquisitions in five years, spending less than $100 million in cash and stock, the M&A strategy played a vital role in transforming Opsware from a single-product player, solely dependent on one giant customer, to a full-suite data center automation market leader with hundreds of enterprise customers.  The acquisitions positioned the company for a major strategic partnership with a technology giant in 2006 and a billion-and-a-half dollar sale to another a year later.

Who you gonna call?

Just like the Loudcloud sale to EDS, Opsware’s M&A strategy relied on many essential ingredients and actors: an aggressive but disciplined management team, outstanding engineering, marketing, sales, legal and finance functions, an acquisition currency and a supportive board.  Again, however, a small effective business development team was there to leverage these assets and drive the process to achieve results while the rest of the company continued to fire on all cylinders.  While we acquired four companies, we evaluated hundreds and gained massive industry insights in the process.

What’s your M&A strategy?  Who’s driving it?

Next up: Strategic Distribution