This article appeared in the Harvard Business Review. It can be found at Start-Ups That Last.
Why do so many start-ups that seem to have it all—customers, cash, a promising outlook—run off the rails? Ask a venture capitalist, and you’ll probably hear that they have trouble “scaling.”
What does that mean, though? VCs typically describe it as a need to “professionalize the organization” and “bring in grown-ups.” But those are simplistic fixes—poor substitutes for the substantive changes that need to occur. Start-ups these days grow so rapidly that it’s difficult for them to correct course once they recognize missteps. They can improve their prospects by understanding the mechanics of effective scaling before they reach that moment of truth.
Venture capitalist Ben Horowitz compares scaling to a “black art.” He and others have proposed useful ideas for demystifying it, but start-ups still lack a cogent framework for transitioning to mature firms. That’s what this article provides. Drawing on our extensive case studies of fast-growing companies and on 75 years of organizational research, we have identified four critical activities for successfully scaling a venture. Firms must hire functional experts to take the enterprise to the next level, add management structures to accommodate increased head count while maintaining informal ties across the organization, build planning and forecasting capabilities, and spell out and reinforce the cultural values that will sustain the business.
It’s easy to misconstrue these activities as replication—as merely increasing the capacity and efficiency of what you’re already doing. But they’re also about handling greater market and organizational complexity as you seek different avenues for growth. That can mean developing new products or services, entering new markets, or engaging in other forms of innovation.
Many entrepreneurs will resist these activities. They often develop strategies opportunistically, lacking a frame of reference because they are starting from scratch, and they take a similar ad hoc approach to building their organizations. Founders tend to view formal structures and processes—elements common to all four activities—as bureaucratic threats to their entrepreneurial souls. They also worry about losing speed, control, and team intimacy. When they eschew order and discipline, however, they pay a steep price: chaotic operations and unpredictable performance.
Scaling doesn’t mean that ventures should disavow their start-up identities and embrace large-company dogma once they’re poised for growth. But those prepared to manage that growth—and to learn new ways of operating and behaving—stand a much better chance of making it in the long term.
Defining Specialized Roles
Founders typically do a bit of everything—basically, whatever it takes to get the business off the ground. Through informal channels they hire fellow generalists, who cobble together their roles and responsibilities partly by pursuing their own passions and partly by looking around and seeing what needs to be done. This idiosyncratic “all hands on deck” approach can work fine in the beginning, when adrenaline is high and the company is small. But as organizations expand, they face new levels of complexity that require them to define and assign tasks more formally.
To accomplish this, they typically seek specialization in select functions, such as sales, human resources, marketing, R&D, and manufacturing. This benefits them in two ways. First, the specialists use their knowledge to tackle their functions’ work more efficiently. Second, as they introduce and implement best practices within their domains, they catalyze future growth by creating slack in the rest of the firm. People who no longer have to worry about marketing, for example, are free to explore other activities.
Of course, all this can create tension between the “old guard” generalists and the domain experts. Demands for functional expertise often outstrip early employees’ abilities to keep up through organic learning. As a consequence, functional leadership titles increasingly go to outsiders, and the legacy folks may grow resentful. Early employees may also chafe against the narrowing confines of their changing roles. Not every generalist can or even wants to become a specialist. Often people get frustrated and leave, taking their valuable relationships and their tacit understanding of the firm’s mission and culture with them.
To keep people working together constructively, it’s important to anticipate and manage these growing pains. Let’s look at how one start-up, Birchbox, did so.
Birchbox experienced explosive growth within just a few years of its founding, thanks to a business model crafted around consumers’ discoveries of new beauty products. Each month subscribers received a box of samples customized according to their personal profiles. They could simply pay their fees and enjoy the samples; they could also go to the Birchbox website and buy larger quantities of the products they liked most. A dedicated team generated a steady stream of digital articles and video tutorials about beauty trends to further engage customers. This model attracted a million subscribers in the first four years, inspiring dozens of copycat start-ups to pitch their businesses as “the Birchbox for X.”
To keep up with demand, Birchbox grew from eight employees in 2010 to more than 300 in April 2014, when it secured $60 million in series B funding. In the process, employees’ roles and responsibilities shifted. Nicole Fealey, the director of people operations and performance, recalls the excitement of being a jack-of-all-trades during the first 18 months. “That’s what I love about start-ups,” she says. “You never get bored.” But she realizes that she and other early employees lacked the knowledge and experience to handle everything on their own as the company grew—and that they would have burned out if they’d tried.
Consider the logistics of shipping a million boxes of unique samples each month—or the job of building sales relationships with enough partner organizations to continually fill those boxes with fresh, interesting products. To manage such complex work, Birchbox divided it into specialized functions and sought out domain experts to improve the effectiveness of each one. The new hires included a CTO with a computer science PhD from Carnegie Mellon and a vice president of brand campaigns who had been a principal at Booz & Company.
“When I walked in and looked objectively at certain monthly processes, I saw that they had been established in a hacked-together way,” says Kate Price, who served as VP of brand campaigns for about three years before becoming VP of Canada. “My consultant mind immediately went to thinking that we should fix all those things, but I learned pretty quickly to respect the people who at the age of 24 had built a process that was part of the engine keeping the company running.” Cofounder Katia Beauchamp agrees about the importance of appreciating the old guard—a group the cofounders see as essential to Birchbox’s “special sauce.” She says, “I think we do a really good job of showing people how valuable their skill sets are and celebrating the fact that we wouldn’t be here without their collective capabilities.” That attitude has kept early employees feeling valued and engaged.
Even so, they have sometimes struggled to find their place in the growing organization. “It’s scary for sure,” Beauchamp acknowledges. “I don’t think it’s easy for me to this day; I don’t think it’s easy for anybody.” Some people had to hire their own bosses to supervise activities they themselves had nurtured since the beginning, rewriting their own job descriptions accordingly. Matt Field, a former early employee who headed international operations during the big growth phase, saw that as an opportunity for personal development. “I knew I did not have the background or knowledge to take Birchbox to the level of aspiration we had,” he says. “I hired someone who could teach me and empower me to get better at my job.”
Cultivating a learning mindset among employees was key, as was reminding them of the challenges ahead and the ways in which experienced talent could help. Those things got Field and others focused on the greater good of the firm instead of worrying about their relevance and status in the new order. Beauchamp says that she and cofounder Hayley Barna “worked really hard to get people to believe that you can hire people better than you.” Involving members of the old guard in the hiring process assured them they would still have a voice. The founders also talked with them about how the domain experts could mentor them and help them develop their niches in the growing firm.
As more outsiders have joined and settled into functional divisions, early employees have provided cohesion through their broad understanding of how the components of the business model fit together. They also serve as a cultural channel back to the time when Birchbox had no brand cachet—a time when it took great resourcefulness to grab the attention of prospective partners and customers.
“People joke that they could never have gotten their jobs now,” Beauchamp remarks. “The old guard didn’t come in with as much industry experience, but they are superskilled at ‘Birchbox’—at our vision and practices.”
“We worked really hard to get people to believe that you can hire people better than you.”
—Katia Beauchamp, cofounder, Birchbox
Does specialization bring risks? Absolutely. Once functions have independent leaders, employees might hunker down in their silos and stop identifying with the organization as a whole. Tribal instincts can prevent cross-functional idea sharing and innovation, so firms must ensure that informal interactions continue across teams and divisions. When companies are in a high-growth phase, they often forgo relationship-building activities in favor of more-immediate work demands. But over time that can lead to stasis and unoriginality. Firms are better served in the long run by fostering cross-pollination while they organize to support the work that needs to be done. The answer is not to avoid building silos but to find ways of bridging them.
Adding Management Structure
When launching their start-ups, many founders eschew hierarchy because of their egalitarian ideals. But as their firms scale, a growing number of people report to a handful of leaders. Founders may think this allows them to remain in command, because all decisions pass through them. But ironically, their organizations spin out of control as centralized authority becomes a bottleneck that hinders information flow, decision making, and execution. A couple of people at the top can’t effectively supervise everyone’s increasingly specialized day-to-day work; in such a system, accountability for organizational goals gets lost. And employees find it hard to remain focused and engaged when they don’t have managerial guidance and processes. They may become frustrated as they struggle for access to decision makers who are juggling many other projects and people.
That happened early on at CloudFlare, a San Francisco–based start-up that was founded in 2009 and quickly became an important player in content delivery and security for small to medium-size websites. By July 2012 it was serving nearly 500,000 websites, with more than 2 billion daily page views (then about 1% of total internet page views). At that time it recounted some of its growing pains to Harvard Business School’s Tom Eisenmann and Alex Godden, who published a teaching case about it.
In the beginning CloudFlare’s founders proudly—and vocally—proclaimed that they would build a completely flat organization, with no hierarchical titles or HR function. Like many start-up leaders, CEO Matthew Prince wanted to promote flexibility and individual achievement and believed they would be stifled by bureaucratic control. In creating a title-free organization, he also hoped to avoid future organizational chart conflicts, since the people initially heading up the small venture probably wouldn’t be suited to lead a team of 250, and senior roles would inevitably change. “Either the original person gets demoted, in which case he or she will likely leave, or the new person doesn’t get brought in,” Prince said. “Neither is a great outcome.”
Nevertheless, problems cropped up. In the three months ending in July 2012, five of the firm’s 35 employees quit, some citing the lack of a clear midlevel reporting structure and the nonexistent HR practices. They described situations in which they had no one to turn to (short of pestering the founders) if they thought certain practices, such as activities related to software or coding standards, needed to change. Without official policies, they found it difficult to navigate conversations about taking vacation and sick days, balancing work and family expectations, and expensing items. The employee backlash was similar to what Zappos experienced in 2015, when it announced that it was eliminating all titles and managers, and 14% of its workforce—210 people—consequently took buyouts and quit.
“It’s been fascinating to watch people from the engineering team look at the sales team and say, ‘Hey, they actually look happy and productive. Maybe managers aren’t such a bad thing.’”
—Matthew Prince, CEO, CloudFlare
Even though CloudFlare lost fewer people than Zappos, the percentage was about the same. Soon after the departures, Prince acknowledged that the firm needed more structure. “We are under no illusion that these management practices will work forever,” he told Eisenmann and Godden. “You can already see some gaps. People want feedback; they want direction. When we double our current staff, we will need more hierarchy and managers and processes.” A product manager had recently joined the organization, but Prince, who still disliked the word “manager,” called him a “product engineer.” He preferred to think of his employees as being “assisted.”
By 2015 the firm had hired additional managers, along with an HR administrator and a talent recruiter. Prince wanted the organizational chart to remain flexible enough to attract senior hires without discouraging solo contributors, but he recognized that adding management structure was helping CloudFlare grow. When building up some areas, such as enterprise sales, he added hierarchical layers. He recently commented, “It’s been fascinating to watch people from the engineering team look at the sales team and say, ‘Hey, they actually look happy and productive. Maybe managers aren’t such a bad thing.’”
Of course, organizations can take structure too far. Having excess layers in the decision-making chain can slow things down by restricting the flow of information (top-down or bottom-up). It can also demotivate employees by signaling that they’re not trusted to handle their own work. But as we saw at CloudFlare, people find too little guidance demotivating as well.
Firms that complement formal structures with informal mentoring and feedback can keep motivation intact. That’s because those things foster a learning mindset, helping employees grow right along with the organization. Clearly delineated roles and areas of authority also enable people to make faster, smarter decisions locally. They streamline the process, rather than gum it up, and promote individuals’ development. The more decisions people are empowered to make on the ground, the more they learn and the more accountable they become.
Planning and Forecasting with Discipline
Improvisation is integral to young ventures; it’s how they make discoveries. However, as firms grow they need a framework of plans and goals to guide them. That way they can keep trying new things and reacting to dynamic markets, but with an eye toward larger objectives and sustaining the business. Otherwise improvisation essentially amounts to aimless riffing.
Many start-ups, including India’s Micromax Informatics, have learned that the hard way. In 2010 Micromax seemed unstoppable. Having stormed the mobile handset market just a couple of years before, it was selling more than a million units a month. Its four cofounders had ambitions to make the company a global leader, and the numbers seemed to put it on that path: That year revenue more than quadrupled, and net profits more than quintupled. In September Micromax raised $45 million in private equity from Sequoia Capital and other investors, and in October it announced plans to go public.
But in July 2011 the company withdrew its IPO. Its relentless pursuit of growth had come at the expense of business hygiene, and it had lost momentum as a result. Mohit Bhatnagar, a managing director in Sequoia’s New Delhi office, says, “From the outside it looked like a company that had grown exponentially, with great customer adoption of its products. However, on the inside it was chaotic.”
At a board meeting later that year, Micromax committed to major organizational changes. To their credit, the founders agreed to bring in an outside CEO, along with senior leaders from blue-chip firms such as Airtel and HTC. When those leaders arrived, they were struck by the utter lack of planning. For instance, Micromax had done little to standardize market and employee information, let alone use it to inform sales, operations, or talent management decisions.
As the new CEO at the time (he has since left the firm), Deepak Mehrotra led the charge to implement strategic planning. With the founders’ support, he stressed the importance of regular goal-setting and pacing exercises companywide to build a long-term vision. He says, “At my first meeting with my direct reports, in January 2012, I made them write their epitaphs: ‘Imagine that two years hence, all 16 of you are returning from celebrating a great year, and your plane crashes. What would you want as your obituary?’” That was his way of getting managers to think more concretely about the company’s future and set clearer performance targets—things the founders had avoided in their excitement to pursue new opportunities and their reluctance to admit when things weren’t panning out. For example, the founders had initiated aggressive expansions to Brazil and Dubai, undeterred by their limited knowledge of customer preferences in those markets. Once systematic planning got under way, the company shut down those operations.
Micromax also began to bridge planning gaps at the operational and tactical levels. In many functions, managers lacked real-time data. Sales was a prime example: Once handsets shipped to distribution channel partners, the firm had long waits before finding out which models had sold, so placing advance orders with suppliers involved a lot of guesswork. That led to underavailability of fast-moving products and excessive returns of others. It also made it hard to know how much inventory to retain. As a result, the company experienced cash flow challenges and had a limited ability to launch new products until it reached credit and inventory settlements with suppliers.
Cofounder Sumeet Kumar designed a solution: a tool that tracked each phone from shipment by the manufacturer to activation by the user. “I now know at exactly what stage in the supply chain a device is,” he explains. “I know, in the top 20 cities in the country and street by street, the models I have sold.” Micromax completed the tool’s rollout in November 2012. Sales and inventory planning have since become much more precise, enabling the firm to learn within 30 days whether a product “is going to be a rock star or a failure.” This has reduced problems with stock-outs, returns, and cash flow.
Before Micromax applied this sort of discipline, it had an ad hoc style of pouncing on opportunities as they arose, using a combination of tacit knowledge and off-the-cuff fixes. Leaders rationalized this approach on the grounds that decisions had to be made quickly because rivals were close on the company’s heels, looking to copy existing products. In the frenzy to tackle pressing challenges, the tasks of documenting solutions and analyzing how they might have been reached more efficiently often fell by the wayside. People had little interest in establishing routines to deal with repeat issues. When they came up with effective solutions, they rarely shared them companywide; each unit had to discover its own best way of doing things. And when key people left, their knowledge walked out the door with them.
“I now know at exactly what stage in the supply chain a device is. I know, in the top 20 cities in the country and street by street, the models I have sold.”
—Sumeet Kumar, cofounder, Micromax
As Micromax’s leaders discovered, even in a fast-paced, high-growth environment, it’s important to set aside time to plan and to identify and share best practices. It’s easy to assume that such activities are incompatible with agility and managerial discretion. To be sure, overly rigid planning processes can provoke battles over limited resources, which may hamstring innovation. But it’s possible to have freedom within a framework. Setting clear goals and guidelines, systematically gathering and sharing information to shed light on performance and enable better forecasting, and creating processes instead of relying on key individuals to craft one-off solutions—all these promote efficient, smart decisions, especially when the world around you is in flux.
With these interventions, Micromax regained its footing in the mobile market. Its 2015 fiscal year revenue was almost $2 billion.
Sustaining the Culture
Culture is typically a big part of what draws people to join start-ups—and what keeps them going. As employees battle the odds to turn a fledging business into a viable company, working late nights and weekends to make it happen, they’re motivated by camaraderie and a sense of belonging to something important.
Founders recognize how powerful this is and rely on nostalgic, almost mythic, stories about the organization’s first days to get everyone to embrace the culture. That can work while a venture is small and all the employees can personally relate to those stories—but as more people come aboard, leaders may struggle to maintain a strong organizational culture. That’s a problem, because culture may be most important during periods of growth. As a venture starts to formalize its functions and reporting chains, identifying with the larger organization helps employees work across boundaries and engage in the spontaneous collaboration and exchange of ideas the company needs to innovate.
Although founders of fast-growing firms say they worry about losing their organizational culture, few take steps to codify and reinforce it. Their attention quickly shifts to things that feel more urgent, such as operations and marketing. As a result, employees’ motivation and engagement slip and people leave, hoping to recapture the magic somewhere else.
How can entrepreneurs prevent these consequences? They can start by clearly articulating their cultural values in their mission and vision statements and in job descriptions. That makes it easier to recognize cultural drift before it goes too far. It also helps the organization keep its values alive by hiring for cultural fit and rewarding desired behaviors through recognition and compensation.
“Most of the personalities here talk a little faster. I tend to hire people who inherently have a bit of discontent.”
—Ryan Howard, cofounder, Practice Fusion
Let’s look at how this played out at Practice Fusion, a San Francisco firm that makes a cloud-based platform for electronic health care records. By late 2013 it had hired nearly 200 employees within 12 months, more than doubling in size. Throughout that period, cofounder Ryan Howard made it a priority to preserve the organizational culture.
One of the firm’s tenets, “Be scrappy,” harked back to the days when the cofounders, spurned by VC investors, worked out of coffee shops and used insurance money from a motorcycle accident to cover payroll. So it’s not surprising that early on, the leaders relied heavily on folklore—tales about their marathon workweeks and bootstrap solutions—to convey this core value. But as the business became larger and more complex, that created distance between leaders and employees. The founders’ charisma and stories were no longer enough to bind everyone together.
Anticipating this problem, Howard articulated the firm’s values more formally, posting them online and in the building. You’ll now find them painted on the office walls—not just scrappiness (his personal favorite) but also integrity, customer focus, teamwork, fun, community giving, and “doing extraordinary things.” These values have become criteria for hiring and performance evaluation. For example, leaders expect employees to be resourceful, self-motivated problem solvers. “Most of the personalities here talk a little faster,” Howard says. “I tend to hire people who inherently have a bit of discontent.”
The firm also instituted weekly town-hall-style meetings, where the founders encourage employees to ask tough questions about what matters most at Practice Fusion, the problems it faces, how key decisions were made, and so on. This not only ensures a regular line of communication with employees but also makes everyone insiders, privy to the leaders’ thinking about critical issues.
The staff also comes together once a month for “phenomenal Friday,” where divisions take turns sharing updates and challenges. Sitting together and swapping stories helps weave the different groups into a unified community—an effort that requires regular care and feeding, especially in a company that extols scrappy self-starters.
As just about any rapidly growing start-up will attest, scaling up is challenging. Market crashes, unreliable supply and distribution partners, fierce rivals, and plenty of other external forces can buffet firms. But that doesn’t mean companies need to be chaotic on the inside. Effective internal organization frees them up to keep pursuing new opportunities and brings long-term survival within reach.
Entrepreneurs may worry that the changes we propose will be the death of spontaneity, adaptability, and speed—everything that got them up and running in the first place. Indeed, these are valuable qualities. Many large companies realize that too; that’s why they often try to behave more like new ventures. We’re not suggesting that start-ups abandon what made them special and innovative. But it’s a lot easier to launch your rocket ship in search of new horizons when you don’t have to worry that someone forgot to fill the tank.
Between the extremes of ad hoc and prescriptive organizing, there’s a useful middle ground. Leaders who can find it will have an edge on their rivals—and that really matters, given how few new ventures become established players.