Why build an Inside Sales Lab in your own office? Testing is an excellent way to grow sales. Most firms with significant inside sales already test alternative scripts, lists, and/or prices, but they do so casually as part of day-to-day activities. A lab concentrates this function into a small, well-disciplined group with the right tools to capture subtle improvements. Take it from today’s guest blogger, inside sales expert Jeffrey Feuer of The InsideSalesLab. He’s got some tips you’ll want to know about.
1. Assess your product or service with a fresh and open mind. Defining limits – in this case to your product – is the very essence of testing. A good test neither limits testing to the current configuration nor stretches into configurations that cannot be realistically supported at scale.
First, think creatively. What terms, features and prices do my prospects crave? Then think practically. What can I deliver with reasonable profit? Test versions that fit both questions.
An ideal test will include a range of terms, features, and prices that fully measure the marketplace.
2. Build a test matrix. Because testing resources are finite, testing every combination of terms, features, prices, and audiences is impractical. Only reasonable combinations can be tested.
Prior to the first dial, create a solid, written road map of the desired tests in order. Create contingency plans to have at the ready when real-world results become available.
3. Imitate the roll out. In your test, include only design elements you can reproduce at scale. This means selecting agents, leads, software, and other factors that are widely available.
Testing is most rewarding when followed by a successful roll out.
4. Welcome (some) failure. Failure is a natural and desired element of multi-celled testing. Failure tells us we have reached the limit along a particular dimension. For example one should generally test prices starting low and moving high until a failed price point is reached. The failure tells us we have maximized potential revenue.
Failure also provides valuable feedback and allows good adjustment to future test cells. Of course we don’t want every test cell to fail but a few failed cells are healthy and inevitable.
5. Randomize your leads. To be of maximum value, every test cell must be homogeneous except for a single test variable. This includes leads. So while it may be good practice to match leads with agent strengths (such as geography or product specialty) in normal operations, the practice is sub- optimal in a testing environment.
6. Recruit representative sales persons. For a test to be readable, the agents conducting it should be neither super-stars nor laggards. The same agents should be used in every test cell.
7. Instill discipline. Most sales agents do whatever they ethically can to close a sale. Sales managers appreciate and cultivate this desirable trait.
In a testing environment this trait is desirable too provided the agent remains within the confines of the test cell. When testing, “sticking to the script” is more important than closing another sale. Some salespersons have a hard time limiting themselves in this way.
For example, if we are testing free trial vs. lower price, our testing agents must always offer only the free trial or only the lower price in their respective cells. Crossing offers or, worse combining them will invalidate the results.
8. Monitor for compliance. Testing only works when the test plan is followed. The only way to assure this occurs is to monitor calls.
9. Act nimbly. As a test progresses, use the results from early test cells to alter future test cells. You should adjust, eliminate, and add test cells as data from earlier test cells becomes available.
10. Be significant. Avoid the temptation to prematurely judge and end a test. This applies to early success or failure. To assure meaningful results each test cell must be performed a statistically significant number of times.
For example early success in an important test cell might tempt one to move onto another test cell. If however the early success does not represent a true result, the roll out will disappoint. Conversely early failure leading to early termination of a test could prevent a successful concept from reaching fruition in the marketplace.