The idea that enterprises can act more like startups to drive transformative outcomes may not be brand new but it’s especially important in today’s software-defined world. Eric Ries and Steve Blank have popularized the lean concepts that form the root philosophy of most successful startups.
As someone who has helped deliver 1,100+ software projects at Gigster and interacted with CIOs and CTOs of some of the world’s largest companies, I have seen these ideas adapted and evangelized within enterprises, not just startups, to help them achieve desired outcomes. Despite these efforts by large companies to adopt the agility and nimbleness of startups, most enterprises are, in practice, far from this ideal.
To succeed, it’s critical to identify exactly what it means to act like a startup and how to overcome challenges that are more likely to arise inside of large organizations. Drawing from experience with hundreds of organizations, I will review core startup principles and then share five recommendations that address enterprise-specific challenges to driving startup-like speed and outcomes.
The underlying motive of a startup is to build something customers want and to obtain those customers in a scalable and profitable way. Good startups want to understand customer needs, build to those needs, and acquire more customers.
Great startups also do this as quickly and cheaply as possible.
To act like a start-up, enterprises should focus on these five core principles:
The invariant is not your idea; rather it is what customers want. Great startups will even change their idea completely if that’s what’s necessary to wow customers. Kevin Systrom and Mike Krieger, inventors of a location check-in app, realized that people only wanted to use the photo functionality. They listened and the result was Instagram.
Great startups understand that every product decision is an experiment to test a hypothesis about what the user needs. If we think users will invite their friends, we could simply add a fake invite button to the app and track how many people clicked. Avoid waste by getting creative about how to test assumptions.
Determine your goals and the metrics you will use to measure success.
For example, if you would like to drive great customer experiences (a goal), then the metric might be your churn rate. As the scale of users increases, inexpensive surveys and analytics tools exist to measure what’s important, understand the impact of experiments, and determine your next investment.
If a startup is solving a critical problem with no existing solution, users will be eager to try an early-stage product. The implication? Solve the problem with the least amount of effort possible. This is the idea behind the famed Minimum Viable Product or MVP. When we build something concrete and get feedback from customers that it is valuable, we are emboldened to build the right next product and maximize our odds of success.
Time to market is critical and can be the difference between staying alive and going bankrupt. The purpose of speed is to maximize the rate of learning. Great startups don’t move in a straight line—they zig-zag around a market landscape based on what they learn and iterate toward whatever direction is optimal for customers.
Acting like Startups By definition, an enterprise is large. Size is the biggest hurdle in successfully emulating the agility of a startup. As N goes from ~5 to 5000, nearly everything becomes larger and non-linearly more complicated. Coordination costs, communication overhead, code bases, project sizes, budgets, and the cost of reskilling employees all increase with company size. What to do? Go small. It’s not a small idea.
As enterprises tackle digital transformation challenges, thinking small in the scope of the project and the size of the team, and thinking frugally in the use of resources, can actually have a large business impact. Here are five paths I recommend to help enterprises function like a startup:
A study of 50,000 projects by the Standish Group showed that small projects are 10-30 times more likely to succeed than large ones. An Oxford University study of 1,427 IT projects found that one in six projects cost triple the original estimate and highlights a $5M project that ballooned to $200M. Stories like this are not uncommon. While enterprises have the means to work on really large initiatives — and often must when they are refactoring underlying infrastructure — the startup-like approach is to break big projects into smaller self-contained ones and iterate towards the promised land.
Businesses evolve and increase in size in order to reduce the coordination costs required to get larger amounts of work done. Unfortunately, with increasing process and bureaucracy, they lose the nimbleness that smaller groups enjoy. To maintain their agility, enterprises can ensure new projects are worked on by smaller groups of ~ 5 people. Jeff Bezos is said to have instituted a “two pizza rule” in the early days at Amazon, effectively restricting team sizes and meetings to groups that are no larger than a number that can be fed by two pizza orders.
Product management is a critical skill that ties the needs of the users and the goals of the business to the products being developed. The product manager role is often conflated with that of a taskmaster who drives projects to completion and moves around tickets in the sprint. Disciplined project management is crucial, but strong product management is the key to ensure that business impact and user value remain the north star.
In addition to product management, it is critical to have a team that has an up-to-date understanding of what relevant APIs, open source and proprietary tools already exist. The best teams use existing tools to build minimally and experiment quickly. In large organizations tasked with maintaining mission-critical legacy systems, it can be cumbersome to re-skill teams and learn cutting-edge technologies. But to succeed at working small, large companies should remove incentives to build proprietary tools and instead reward teams for taking advantage of open source alternatives as a means to accelerate the path to working solutions.
Find or create teams within the organization that exhibit startup characteristics–ones that are dynamic and entrepreneurial. Direct them to mentor new agile teams within the company. Alternatively, look outside the company and work directly with small external teams that already act like startups. But be sure there is a sufficient flow of knowledge to ensure a gradual build-up of internal capability. Protect these “internal startups” from being smothered by standard enterprise processes to give them a chance to get big enough to influence the mother ship.
To summarize, large enterprises should start small and effect change from the bottom up. Pick an important goal, and start with a small project and a small team with the charter of proving key hypotheses. Ensure that strong product management skills are present and that the team is aware of existing and upcoming solutions that can provide leverage.
Finally, nurture entrepreneurial teams by complementing your organization with external entrepreneurial teams that already have startup DNA, or build from within and shelter them to allow them to grow. In a software-defined world, bigger is usually not better. WhatsApp had 55 employees when it was acquired for $19B (or $350m per employee), Instagram had about 13 employees when it was acquired for $1B (or $77m per employee) and Bitcoin has $67B worth of cryptocurrency in circulation and has zero employees.
Enterprises have a tremendous amount to gain by adopting the same mindset that birthed these successes. In software, leverage is the name of the game and the big winners start small.