How to Use This in GLIDR

Off-Brand Testing is a technique where well-established companies launch a new product or service under a different brand so as not to skew customer impressions with their current brand name and all of its associations. 

In GLIDR, this test is a relatively large one since you are launching a new product or service. Since there is a defined result you want to get -- a measurable level of traction or customer interest -- you can run the test as an Experiment. First, in the Plan phase, define what success will look like and why you want to do this particular test. Move it to Run when you start to roll out the product to users, and measure your results as they come in with Evidence - Other. When you've reached the end of your test, move the Experiment to Analyze and figure out if you want to grow the brand under this name, include it in part of your current product lineup, kill the product, or something else. 

Learn more about each of those aspects of GLIDR:

Article excerpted from The Real Startup Book
​

Off-brand testing is a technique that helps established companies run experiments while minimizing risk to their brand equity. Off-brand testing prevents skewing the customers' response (up or down) due to their brand. In a nutshell, the product or service being evaluated is presented using a no-name brand or without branding at all. Typically, these are executed as very small scale tests in order to get a statistically meaningful result. If it comes back positive, the data justifies a larger batch of effort.

While the numbers vary based on product and market type as well as survey, most customers need 8-21 exposures on average to a brand message before they'd consider purchasing a product. As these numbers are typically quoted as averages, they presumably describe the behavior of the majority, not of the early adopters. With repeated exposures to a specific message, customers start to associate certain emotions or thoughts with a specific product or company. By increasing the number of exposures, a marketing team can create a familiarity with the product, to at least consider it when purchasing a product from that category. When evaluating branding initiatives, the key metric is the number of exposures a consumer has to the core message behind the brand.

In truth, most "businesses earn almost all their profits from a small number of brands — smaller than even the 80/20 rule of thumb suggests. In reality, many corporations generate 80 percent to 90 percent of their profits from fewer than 20 percent of the brands they sell, while they lose money or barely break even on many of the other brands in their portfolios," according to Nirmalya Kumar writing for the Harvard Business Review.

With off-brand testing, you are essentially trying to discover which early adopters are the most likely to respond to an offer (without adding the persuasive influence of a known brand to the mix). This is trying to measure the level of passion or pain the particular group is feeling, with respect to this problem or your solution. So the key unit of measurement is typically the conversion rate. If you show this product, for example on a landing page, to 1000 ideal prospects, what percent take action?

If these customers have been repeatedly exposed to the brand, particularly with positive associations, the conversion rate obtained will likely be higher than if it was not related to the brand. It may even skew results enough, to give you a "false positive." You walk away from a test thinking you have a good product idea, but in fact you obtained a positive result because the customer knows and trusts your brand.

Alternatively, if a company or brand has a bad association within a particular segment, it can use off-brand testing to anonymously evaluate products or features.

Another common application of this approach is to launch an app in an app store in a similar country to your final target market. Tongue-in-cheek, this technique is referred to in Silicon Valley as "test in Canada."

Negligible, as this is done alongside other techniques mentioned in the startup realbook. For example, on a landing page smoke test, you'd avoid using corporate logos, color, or fonts that could hint at the identity of the experimenter.

Off-brand testing is similar to:

As this is a technique used to reduce bias, there are no major additional biases. It's worth noting that you can still go overboard with selling instead of validating, even without branding. There are a number of techniques that will influence the outcome but will not necessarily tell you about true demand. For example, heavy scarcity tactics will increase "sales" but not help understand the underlying demand. On the flip side, having too many options will depress response.