In contrast to "corporate research," sectoral investigation includes research on the industry as a whole and the firms that make up the industry.
Corporate research starts and ends with a focus on a particular company as a target. Ultimately, good corporate research would include all the elements of sectoral research. And sectoral research, when it gets to the application phase, could require the intense close-up, even microscopic, focus on a company, or set of companies.
Take care to note the distinction among industry-wide characterizations, subsector or regional characterizations, and firm specific descriptions.
Reports should not confuse generalizations about the industry nationwide and assume they apply to each firm (ecological, or aggregate, fallacy) nor should findings based on a specific firm (or a few firms) be assumed to represent the whole industry (atomistic, or individualistic fallacy).
Be sensitive to the history and evolution of the industry.
The present cannot be understood without reference to past: trends and key turning points draw attention to the dynamics of industries.
Anticipate that data about industries will have "holes" due to factors such as reporting lags, data collection costs, and confidentiality restrictions.
This means that it is frequently necessary to "work around" missing data by interpolating, aggregating, or living with the gaps.
Quantitative data analysis must be balanced with qualitative data.
The analysis must be grounded in—and tested by—experience and direct inquiry. Contextualize quantitative findings. Make full use of published reports, student papers and theses, trade associations, the press, knowledgeable experts, unions, and personal experience in addition to carrying out field research—which includes site visits, interviewing and informed observation.
Provenance is essential.
Know where the data come from and keep careful records of sources. Don't make assumptions, always question, strive to understand the sources and their methodology, and always attempt to cross-check your data. Don't rely on official data reported in secondary sources. Check out cited sources.