Over the past year, ICIC has partnered with JPMorgan Chase to research the opportunities and challenges associated with incubators and accelerators. The lack of data is a major barrier, making it difficult to benchmark their performance. In this blog we share our approach and new data that we collected from eight high-tech incubators and accelerators in the U.S. as a starting point to address this challenge.

There are a number of measures that can be used in assessing the performance of incubators and accelerators. The International Business Innovation Association (InBIA), for example, offers an incubator self-assessment tool to help incubator and accelerator managers evaluate their organization’s performance across ten programming areas. Other performance evaluations focus strictly on the performance of the participating businesses, typically using survival rates, revenue growth and profitability as performance measures. The specific mission and goals of incubators and accelerators are an important component in defining success at these organizations. For example, a non-profit incubator may emphasize local job creation as a key performance metric while a for-profit accelerator may focus on venture capital investment.

In our research, we collected extensive qualitative and quantitative information and developed an approach that analyzes the effectiveness of both the programs and resources offered by the incubator or accelerator as well as the performance of the businesses they support. The following summarizes our approach and provides a few performance measures that should be relevant for most high-tech incubators and accelerators.

Our Sample

The eight high-tech incubators and accelerators in our study support businesses in general technology, biotechnology, cleantech, health IT, water technology and interactive media. Seven provide tenant spaces, while one operates their programs virtually. They are located in different cities spread across the country.

In 2015, ICIC surveyed all of the businesses supported by seven of the incubators and accelerators over the past five years (538 businesses). It was a comprehensive and independent survey. We achieved a 38 percent response rate (206 businesses). One incubator provided their own survey results from an additional 89 businesses.

The surveyed businesses were mostly small, young start-ups. They have on average 11.3 full time equivalent employees (ranging from 3.4 to 31.3), making them on average smaller than the average U.S. business, which has 12.7 full time equivalent employees. They are also younger than the average business in the U.S.: 83 percent of the businesses were 5 years old or younger, versus 35 percent for all U.S. businesses. Nearly 80 percent of the businesses considered themselves to be either seed or early stage businesses.

Program Effectiveness Metrics

We created program effectiveness indicators for the three primary types of support offered by incubators and accelerators: business education, capital access and network connections. Our survey asked the businesses to indicate which resources provided by the incubator or accelerator were effective in helping their companies grow.


KPIs for Start-Ups 

In our survey we collected data on many performance indicators, but offer a summary of five key performance indicators (KPIs): revenue, net income, success rates in accessing capital (equity, debt and grants), amount of each type of capital raised and patents received. Tracking more than revenue and net income is important for measuring performance in tech start-ups. For example, some businesses will have a long runway until they start generating revenue and profit, but in the meantime their progress can be measured by other indicators, such as capital raised or patents received.


Control Group Comparison 

The gold standard for measuring the impact of any intervention is using a control group. To assess the effectiveness of the eight incubators and accelerators we studies, we created a control group of businesses for each organization that did not receive any support from them. The control groups included a set of 2015 Dun & Bradstreet businesses within each incubator or accelerator’s target market that matched the businesses they supported in terms of business age, employment size, and industry. We excluded any businesses that were included in contact lists of the incubators or accelerators we studied. Since the control group businesses may have received support from other incubators or accelerators, we are only measuring the impact of the specific incubator or accelerator in our study and not the effectiveness of these organizations in general.

For six of the eight incubators and accelerators in our study, the businesses they support on average outperformed their control groups in terms of revenue. For four of the incubators and accelerators, their businesses outperformed their control groups in terms of net income. Across all organizations, 13 percent of the businesses they support outperformed their control groups in terms of revenue and 28 percent outperformed in terms of net income. Analyzing the performance differences of the incubated or accelerated businesses and their control groups over time would yield more information about the effectiveness of the incubators and accelerators in our study.


The baseline data presented here can be useful to other high-tech incubators or accelerators looking to benchmark their performance. In addition, we hope our approach and data spurs more robust performance studies of incubators and accelerators. It is important to note that we did not include any community impact measures. Measuring local impact beyond start-ups is a separate but equally important component to our research on incubators and accelerators and is discussed in other blogs.

This post originally appeared on the ICIC blog
Photo Credit: Guido van Nispen via Flickr