4 Times 'Fail Fast, Fail Cheap' Is the Wrong Advice Used at the wrong time or taken to an extreme, failing fast and failing cheap can result in significant destruction of value.

Editor's Note: In the new podcast Masters of Scale, LinkedIn co-founder and Greylock partner Reid Hoffman explores his philosophy on how to scale a business -- and at Entrepreneur.com, entrepreneurs are responding with their own ideas and experiences on our hub. This week, we're discussing Hoffman's theory: If you're not embarrassed by your first product release, you released it too late. Listen to this week's episode here.

When launching a new concept, an unproven idea, it is often more important to get something to market quickly than it is to continue to refine the concept privately. For years, we have advised startups with a new idea to fail fast, fail cheap.

One thing we know is that new ideas will almost certainly need to be changed, modified and tweaked based on the market's reaction before they are ultimately successful. Therefore, don't spend inordinate amounts of time and money developing the details of your concept and adding all of the bells and whistles. Get it to market as quickly as you can. Get feedback from the market. Learn what works and, perhaps more importantly, what doesn't work. Adapt, adjust, innovate and relaunch. You'll reach success much more quickly and at a much lower cost.

Related: Reid Hoffman: When It Comes to Launches, Imperfect Is Perfect

Michael Bronner, the founder of Digitas, also founded a company called Upromise. The premise of Upromise was that companies who sold to consumers (e.g., packaged goods companies, restaurants, etc.) would be willing to rebate a small percentage of the purchase price into an account that their customers could use to fund college education.

Upromise would get paid for providing the platform that would host the accounts, signing up partners who would provide rebates to customers and generating new members who would buy from the partners. For the partners, the benefit was incremental sales. For the members, the benefit was free money. The concept seemed plausible, but the concept was totally unproven.

In the early days of Upromise, the most critical thing was to demonstrate that the concept worked and that the company could be profitable. Many ideas were tried. Those that worked were rolled out. Those that didn't were shut down. There were many failures, but enough successes to demonstrate that the business worked and could be profitable.

Related: 5 Lessons You Learn From Your Business Mistakes

Failing quickly and failing cheaply led to Upromise succeeding and ultimately being sold to Sally Mae. The early investors reaped a considerable profit.

However, while failing fast and failing cheaply is a concept that is valuable in the right circumstance, it's not always appropriate. Four examples of times when this is not the best approach include:

You are launching a proven concept.

When you are launching a new concept or a new idea, getting to market as fast as possible, failing quickly and cheaply, testing and learning, makes a ton of sense. This is not the best move when you are launching a business that is a proven concept -- one that others have done successfully many times. Rather than reinventing the wheel, learn from the experiences of those that went before you. For example, suppose you were launching a McDonald's franchise. The things that have worked in the past are well understood. Take the time to learn from the mistakes and the successes of others who have gone before. You'll get to success much more quickly than you would through trial and error.

The business has built brand equity.

Even if you start with a new concept or idea, you'll need to abandon the fail fast, fail cheap approach once you have had some measure of success. When a business is first starting, it has no equity on the balance sheet nor does it have any brand equity in the market. At this point, there is very little cost to trying and failing. The business literally has nothing to lose. However, once a business has experienced success, this is no longer the case. It does have something to lose.

Related: To Keep Stores Fresh, This Franchise Looks Within

Once a company has launched a successful product or service, it needs to be careful about introducing a new product or service that might fail and in so doing damage the brand equity it has created. At this point, failure can destroy previously created value. Consider a clothing designer who has released several successful lines. New releases that are significant failures could hurt the way customers view the brand.

Failing fast and cheap, learning from your mistakes and moving on is great when you have nothing to lose. Once there is equity on the balance sheet and brand equity in the market, more care must be taken. At this point, failure of a new offering could destroy brand equity which might, in turn, destroy value on the balance sheet.

The business has grown and needs infrastructure and stability.

Tactics must also change as the complexity of the business increases. In the early days, Upromise was introducing new ideas at a furious pace -- faster than operations could keep up. By the time the concept had been proven, the back office was being held together by bubble gum and bailing wire.

For example, one of the critical pieces of information on the Upromise website was the amount of the rebate for any given product. In fact, this piece of information might be listed in a dozen places on the website for a single product.

It would have made complete sense to have had a master location that contained this information for every item on which Upromise offered a rebate. Every other place in which this information was shown would be a variable that simply pointed back to the master location. In this way, if the information in the master location was changed the data in every other location would change automatically.

Related: 5 Times Bill Gates Screwed Up

Unfortunately, the website wasn't constructed in this way. In order to get new products and concepts launched in the quickest way possible, every location in which this data appeared contained hard wired text that had to be changed individually. Sadly, there wasn't even an index that spelled out where each instance of this critical data was located on the website. Making a change to this data meant conducting a search to find all of the locations where it existed. Not surprisingly, errors were not an infrequent occurrence.

Once the Upromise concept was proved, management finally had to stop the onslaught of new idea implementations long enough to fix the infrastructure. Subsequent changes had to be integrated into the new infrastructure. It took a bit longer but enabled operational stability. Upromise was certainly right to prove its concept as quickly as possible. There really was no choice. However, at some point discipline had to be introduced.

Another example of this is the genesis of Capital One's credit card business. As a division of Signet Bank and later as a new spin-off, the company was very lean. There was little in the way of infrastructure or bureaucracy to slow the company down. Innovations were frequent and came in rapid succession.

For example, Capital One introduced a new concept, the balance transfer. Initially, this program was launched on the PC of one of the IT staff members. Another innovation was Capital One's approach for profitably serving customers whose poor credit scores had previously prevented them from obtaining credit cards. A combination of low credit limits and high fees made this previously unserved market highly profitable.

Innovations like these led to explosive growth for Capital One. However, the company finally reached a size where it had to develop the risk infrastructure and therefore the bureaucracy more typically associated with a large bank. In this case, the change was precipitated by a memorandum of understanding issued by the firm's regulators. Unbridled, rapid innovation can be a powerful asset in the early days, but at some point, a larger business needs the infrastructure and the internal checks and balances to ensure stability.

Related: 7 Reasons Your First Business Will Fail

Fail fast, fail cheap, but don't fail because you are cheap.

Even in the early days of a new concept when failing fast and failing cheap is often appropriate, businesses don't want to fail because they are cheap. That is, make sure that the thing you are taking to market is robust enough to have a chance of succeeding. You won't learn much about the viability of a concept if the item that you launch is so cheaply constructed that it breaks as soon as it's taken out of the box.

Used at the right time and in the right way, failing fast and failing cheap is a powerful tool. Used at the wrong time or taken to an extreme, this concept can result in significant destruction of value. Like fire, used appropriately it can be a powerful force for good, used inappropriately it can cause great harm. Choose wisely.