Manipulating and structuring research results can help you to put together seemingly isolated pieces of data and achieve an unexpected result.

Outlines and Templates. How are you assembling the data to help you produce your analysis? One way is to use an outline. The outline format is particularly easy to handle when you are using word-processing software.

• First, create an outline of the topic you are analyzing.

• Then, as you review the raw data, insert each piece of data as many times as needed under every appropriate heading.

• Do this for all your data.

• Then read the outline, section by section. What conclusions do you see? What problems and gaps do you notice?

Data versus Conclusions. Another tactic is to separate the raw data, visually or physically, from your conclusions. By doing that in the draft stages, you can then go back through it and make sure you have enough data to support each of the conclusions you have drawn. If you do not find enough support, you can then add the data you did not include in the right place or drop the conclusions as unsupported. In addition, you should read the analysis to verify that you have drawn conclusions from all the data you have included in it. If not, ask yourself why you included each piece of data in the first place.

Coverage. Typically, you will find that the raw data you collected on a competitor will fall into in one of three basic categories:

• what your competitor says about itself

• what one competitor says about another competitor

• what third parties say about your competitor

Inferences. When you study raw data and try to come to a conclusion about what it all means, one tool you should use is to try draw inferences. That involves coming to a conclusion in light of both logic and your own past experience. However, that same process can mean that you fit incoming data into your own preexisting beliefs or even see what you expect to find. In other words, your own experience can act as a screen on the data as well as an aid in analyzing it correctly.

Just being aware of the difficulty of dealing with inferences can help you avoid the pitfalls. Use this simple test to see whether you are having a problem dealing with inferences. Ask yourself, as each new piece of data comes in, which of the following is your immediate reaction: "That fact is incorrect (or correct)," or "That fact must be incorrect." If your response is the second, and not the first, you may be fitting the data into your preexisting beliefs rather than testing it to see what it really means.

Omissions. Do not be afraid to say that you lack enough data on something to reach a valid conclusion. The presence of a gap may be significant, or it may simply represent an area where more work needs to be done. In particular, the presence of a major gap should be an alert of the need to develop or supplement that data.

What is not present after you have finished your research can often be as significant as what is present. For example, you may have found that a competitor is planning to spin off a particular operation. From previous analysis, you may have found that this operation is a highly profitable one. If you can find no reason for the proposed spin-off, you should consider that a significant omission.

Then try to establish what the most plausible reasons might be for this action. In this example, there may be two:

• a possible need by the parent company for cash for its other operations

• a technological breakthrough by a competitor that might soon make this operation less profitable or even obsolete