In my third commentary arising from programming at the Global Entrepreneurship Congress last month, I share comments about evidence to determine the effectiveness of accelerators. While many startup success stories like Facebook did not involve accelerators, most entrepreneurs now consider them part of the start and scale journey. This has driven demand for the launch of hundreds of accelerator programs around the world, prompting us to question how differences across accelerator programs influence startup performance.
Institutions funding economic development have been clamoring to support an explosion of programs that can help growth-oriented new and young businesses scale faster. Working on the assumption that they are beneficial, many governments have been funding and promoting them as part of their efforts to promote economic growth. Accelerators are a classic example of where the field should look beyond the headline news at data for solid evidence justifying investment in these startup support programs.
As with any startup phenomenon, research has not kept pace with the proliferation of accelerator offerings, which vary in their model and formulation in terms of the size and sector of their founder cohorts, amount of seed money provided, level of equity demanded from graduates, length of the accelerator program, location, and learning opportunities offered for budding entrepreneurs.
The good news is that the knowledge gap regarding accelerators is beginning to narrow.
But first, let’s look at why accelerators have found support at almost all levels – federal, regional and local – and across sectors and industries. For example, the U.S. Small Business Administration (SBA) has financially supported accelerator programs via its Growth Accelerator Fund Program. In its first year, the SBA awarded $2.5 million in cash prizes to a group of 50 organizations. The SBA program expanded in 2015, offering $4.4 million to 80 organizations throughout the country.
Why all the hype about accelerators?
Modern-day accelerators emerged in 2005 with the launch of Y-Combinator, followed closely by TechStars in 2006. Shortly thereafter, several other stakeholders decided to follow their approach of a limited duration program for cohorts of early-stage entrepreneurs, with the aim to facilitate connections with potential investors.
Growth in U.S.-based accelerators really took off after 2008, explains Ian Hathaway. “They grew from 16 programs that year to 27 in 2009, and to 49 in 2010, before eventually reaching 170 programs in 2014 and holding mostly steady. All told, the number of American accelerators increased an average of 50 percent each year between 2008 and 2014.”
Beyond economic leaders’ attempts to stimulate entrepreneurial growth, accelerators found strong support within the startup community itself, particularly from venture capitalists (VCs) and angel groups.
Accelerators provide information that business angels and VCs need for diversifying their portfolios of high-potential companies. As explained by Scott Shane, “organizations that typically invest $3 million in a single early-stage venture deal, as venture capitalists do, are not designed to evaluate and assist 120 ventures that each receive $25,000.” Accelerators provide those financiers a service in many ways, through their structures and processes that allow them to make these types of decisions, and push “investor-ready” startups further down the pipeline.
Accelerator graduates have earned a seal of pre-approval when they present themselves to seed-stage investors. The gateway, however, has not gotten significantly wider. On average, members of the Global Accelerator Network (GAN) receive 450 applications and only accept 2.1 percent of them.
What do we know?
Rigorous research on accelerator performance is underway by members of the Global Entrepreneurship Research Network (GERN), a working coalition of institutions seeking to develop a next generation of entrepreneurship research, share lessons and knowledge and establish open, standardized data resources.
The Aspen Network of Development Entrepreneurs (ANDE) and Emory University, both GERN members, announced last year a $2.3 million investment in the Global Accelerator Learning Initiative (GALI), in partnership with a consortium of public and private funders from the GERN network, such as the Argidius Foundation, the Kauffman Foundation, the Omidyar Network, and USAID. The announcement came at a time when there were at least 650 accelerator programs around the world.
Over the course of three years, GALI will be building on the Entrepreneurship Database Program at Emory University, which works directly with accelerator programs to collect and analyze data on how they are structured and which kinds of entrepreneurs they attract and support. GALI surveys all the entrepreneurs who apply to these programs at six-month intervals, including those who are not accepted.
GALI recently completed its first major research report, “What’s Working in Startup Acceleration: Insights from Fifteen Village Capital Programs.” The study compares the performance of ventures that were accelerated against those that applied but were not accepted, based on one-year changes in revenue, employees, and investment.
Overall, the study found that businesses that went through an accelerator raised almost eight times the investment money than non-accelerated businesses. More importantly, however, are the differences between high and low-performing Village Capital programs. Based on these differences, the report authors tested predictions about which programmatic elements impact venture performance within the Village Capital model. Here are some of the features of high-performing programs that emerged from the study:
- High-performing programs planned for more time for entrepreneurs to work on their own, rather than spending as much time as possible delivering program content.
- While high-performing programs had smaller applicant pools on average, their applicants tended to have more intellectual property and more educational, entrepreneurial and senior management experience.
- High-performing programs selected ventures that were younger on average (1.73 years), compared to ventures in low-performing programs (2.47 years).
- High-performing programs put less emphasis on finance, accounting, or formal business planning. They spent more time working on presentation and communication skills, networking, and organization structure and design.
It is important to note that the most successful of Village Capital’s programs were in the United States and Mexico, whereas the lowest performing were in developing countries, which could mean there are other factors in a respective ecosystem that influence the local accelerator approach, as well as the performance of its graduates.
What don’t we know yet?
While the GALI report is far from conclusive, it begins to shed light on the features that differentiate impactful accelerators, putting myths and assumptions to the test. The study is the first in a series of annual reports to be published based on data emerging from the Global Accelerator Learning Initiative. Cuts of data will also be available for use by academic and practitioner researchers in external research projects, further spreading the pace of analysis on the subject.
As research in this field advances, it would be beneficial to set widely agreed-upon criteria on what it means to be an accelerator. There continues to be considerable ambiguity about what qualifies as one. For example, for the purposes of the most recent SBA Growth Accelerators competition, accelerators, incubators, co-working startup communities, shared tinker-spaces, or other models to accomplish similar goals, all qualified.
A Brookings Institution study reviewed nearly 700 U.S.-based organizations that were categorized as an “accelerator” or “accelerator/incubator,” either through self-identification or through the identification in various databases (Pitchbook, Seed-DB, Global Accelerator Network, and Accelerate). Based on this definition of an accelerator, (“a fixed-term, cohort-based program, including mentorship and educational components, that culminates in a public pitch event or demo day”), fewer than one-third of such programs could be said to fit the definition.
While concepts vary between an accelerator versus an incubator, or between an accelerator and a seed fund, there is still confusion among entrepreneurs and support institutions about the differences. Moreover, several hybrids have emerged. Which model of support is most effective might depend on the type of industry being disrupted.
There are many angles from which we should evaluate accelerators as we determine what works and what doesn’t. For example, when it comes to gender, we don’t yet know whether having female mentors and leaders working with the entrepreneurs in accelerators have an influence on the success of women accelerator graduates.
Nonetheless, GALI has taken a decisive step in the right direction. If research institutes continue to align their research agendas to address knowledge gaps in accelerator impact, and accelerators continue to contribute information, GALI data on accelerators will grow and there will be an even wider range of questions that we will be able to answer. This will help all of us improve our accelerator models, design smarter public policy and, most importantly, enable more entrepreneurial growth.
Photo Credit: Flickr
Via Kauffman Foundation's Policy Dialogue on Entrepreneurship.