Beating Business Failure Statistics

Beating Business Failure Statistics

We’ve all heard it—and it sounds pretty convincing and terrifying—80% of small businesses fail in the first 1-2 years. Recently, a lot of business blogs have asserted that this statistic does not have any factual basis. They point to the Bureau of Labor Statistics’ Business Employment Dynamics reports which clearly show that the outlook is not nearly that dire for starting a business.

In fact, data from the Bureau of Labor Statistics shows almost the opposite to be true—only 20% of businesses fail in their first year, some 80% survive the first year, and 50% make it to year 5. They found that about 80% of businesses with employees will survive their first year in business, 70% of businesses with employees will survive their second year; 50% of businesses with employees will survive their fifth year in business; and about 30% of businesses with employees will survive their 10th year in business. That’s certainly better than the 80% failure rate that we so often hear, but don’t jump into the deep end of the pool with the idea that 4 out of 5 businesses succeed and you have always been able to do better than the lowest 20%.

One of my favorite people, Russell Davis (probably quoting someone else) used to remind me, “Torture a number long enough and you can make it say anything.”

It turns out that even the Bureau of Labor statistics are not that clear-cut and may present a dangerous false-positive result. Those statistics are derived from data provided only by companies that are filing payroll reports. Because of the source of the information, the statistics do not take into account the companies that launch without W-2 employees, and our experience indicates there’s a huge number of small business startups that are not included. The truth is that the vast majority of businesses with whom we interact start out with 0 employees. The owner, and often family members, do the work without employees. As they grow, small businesses often rely on contract labor or temporary staff that are actually employed by an agency, not the small business. They are not included in the Bureau of Labor statistics.

The research basically asked, of the businesses that began reporting payroll information 10 years ago (5 years, 2 years, or 1 year ago), how many of them are still reporting payroll? That is a valid way to measure how many employers survive a year or 2 or 5 or 10 years. The tripping points are that some of those “first-year employers” have been in business for a long time prior to that and literally thousands of businesses do not have any employees in their first, second, fifth and even tenth year of business.

We still believe we can find some helpful information from the data that the Bureau of Labor Statistics’ Business Employment Dynamics shared.

Once a business reports having employees, the business, statistically, has an 80% likelihood of surviving that year. A couple of points jump out. First based on our experience, that percentage is higher than we see in businesses that start without employees. There is truth in the adage, “If you don’t have an assistant, you are an assistant.” The common challenge for most entrepreneurs lies in trying to be all things to all people. If you don’t have any employees, you are trying to do all jobs and you have no backup.

However, having employees you can’t pay because you don’t have adequate sales is worse than not having employees. That guarantees business failure at an accelerated rate. This may be clear in the second pattern that jumps out from the Bureau of Labor stats: the longer an employer is in business, the harder it becomes to stay in business; the failure rate goes up each year in business. The takeaway? Spend your time doing what must be done to generate revenue as quickly as possible so that you need and can hire employees. Then, spend your time making sure that every employee brings value to the company that ultimately results in profit.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.