Think of research as an iterative process where you create a chain of searches going from general concepts to specific ones.  You’ll find the right facts when you get down to concepts specific enough to eliminate redundant general content but still broad enough to avoid hitting keyword-stuffed pages like job postings.

The process goes something like this:

1. Search
2. Identify relevant concepts
3. Organized and prioritize them
4. Iterate one level down with searches on the most important topics

The goals are to focus your time on high-priority concepts instead of whatever happens to come up and to start organizing your thoughts for what your deliverable will look like while you are doing research.  The biggest mistake most business analysts make in research projects is to spend too much time looking for data, past the point of diminishing returns, and not enough time thinking about the information and how to analyze it.  Taking a concept-based approach from the beginning is one way to avoid that trap.

A Home Depot example

To walk through a concrete example, let’s say you are doing research on Home Depot’s strategy as part of an analysis of the home improvement retail industry.  One good approach is to start with general background sources to get a sense for the company’s strategy.  In this case, however, let’s try the iterative approach I described, starting with a web search on “Home Depot strategy” to see what comes up.  Walking through the first 30 results or so, I came across a moderate amount of useful information, but more importantly some notable terms to follow up on.  After grouping them together, here are a few of the topics I would look into further:

While some of these company-specific phrases sometimes lead to overly-specific pages like job postings, they can also help you strike the information jackpot.  For example, Home Depot is currently positioning itself as “every day value,” but searching on that term mostly inundates you with job postings and other less than useful pages.  Of course, it also brings up a good presentation on merchandising by EVP Craig Menear, which provides amazingly detailed information on how Home Depot is working to use “destination” product lines to drive sales of complementary products.

Looking at these results, you can start to see the outline of a few analyses that might be interesting to present to a client.  One might be a chart showing Home Depot’s customer satisfaction scores over the last few years combined with changes in staffing policies and other initiatives as well as financial results.  It seems like customer service has been one of the hot-button issues for the company over the last few years, and one that led to former CEO Bob Nardelli’s departure in 2007.

By partitioning your search out this way, you also get halfway to a document outline.  Even when you’re in the middle of looking for data, never let yourself get so absorbed in the research process that you neglect to keep thinking through what you’re going to do with it.

1. Figure out the story

Decide what your story is first.  In other words, what does your audience really need to know?  Let’s say you’re helping to prepare for a sales meeting: emphasize the meeting participants, recent initiatives they may be involved with, and overall priorities for the organization.  Short and sweet.  In contrast, a competitive analysis might entail much more extensive information on sales and marketing strategy, core competences, and profitability over time.  Outline the deliverable at the outset, perhaps in only a handful of bullet points.  Then structure your research framework and related topics around that outline.  If marketing is one aspect, the research should include advertising, product mix, pricing, distribution channels, and a few other concepts like social marketing.

This is going to sound like screenwriting advice more than business research advice, but also think about the story arc.  Two companies in the same industry might have vastly different stories which mean that you should focus on different data points to help explain those developments.  For a struggling company like GM, what are the fatal flaws that led to their problems?  Have they found the keys to redemption, and why or why not?  For a rising star (like Toyota, until recently), what market trends or internal strengths led to their ascendance?

2. Flesh out the background

Use the right background material to understand the target of the research, whether a company, industry, product, or person, before you dive down into detailed topics.  Make sure you have a grasp on the general field, and add concepts to your research framework as you get smarter on the area.  For example, you may realize that platform strategy is an important car industry concept that you might not have planned on looking into initially.

3. Gather information, make connections, and filter

While going through information, focus on avoiding lost time as well as finding the right information.  A lot of inexperienced researchers (or MBAs trained on case studies where the answer is always in the data) spend too much time sifting through mediocre data sources, rather than balancing the data gathering with analysis.  It’s a recipe for poor work products.  You should train yourself to triage a source in the first five minutes and decide how much of your time it merits.  If you’re not making progress at that point, you should move on.  The same is true of using search engines.  While you can sometimes find a valuable data point on the 5^th or 10^th page of results, your time is probably better spent trying a few different, more specific queries rather than just slogging through the results.

4. Analyze

Think about how you can build on the information you’ve found rather than just reporting it.  A quick 2×2 matrix or checklist analysis can help provide some structure and clarity to all the different things most companies are doing every day.  In the car industry, an instructive 2×2 matrix might plot products by list price versus car magazine ratings to assess which parts of the market different companies focus on and are most successful in.  Even simple steps like comparing growth rates or market capitalizations across companies provide context beyond the raw data.

5. Summarize

Boil down the information to as compact a form as possible.  Go for a crisp, short presentation over a data dump.  Your manager or client will not be impressed by seeing every fact you ever came across.  The deliverable’s value will depend on being quickly and easily digested and in providing context and connecting the dots.  That’s where the synthesis comes in.

6. Synthesize

Towards the end of every project, large or small, think about the top three things the audience needs to know.  Typically, they won’t be individual facts, but rather conclusions based on those facts.  Those conclusions should make up the first bullets on a short background document or the executive summary section at the start of longer presentations.  People should be able to get value from your work even if they only have a few minutes to glance at the beginning of the document.

Takeaway

Don’t spend all your time gathering information and giving your boss or client a data dump.  Make sure to filter, analyze, and synthesize.  You should consciously change your mindset as you go through the project in order to be a better researcher and analyst.

What do you think?  What aspects of research projects do you focus on to maximize your productivity?

Why?  First, it’s too easy to end up with irrelevant or only marginally helpful results, especially if you don’t have a set of well-targeted search terms in mind.  Second, it can be very time-consuming to go through search results, and that may impede your progress.  Finally, you never really know when you’re done.  There’s always the temptation to keep searching for that last little nugget of information, and once you do stop, there’s always the nagging worry that you missed something.

So what should you do instead?  Go to more general, comprehensive sources first:

• Wikipedia: Provides good overviews on a lot of topics.  The site is especially useful for getting oriented in a new field you’re not familiar with (when the partner said you should research competitors in the PCB space, did he mean printed circuit boards or polychlorinated biphenyl?) and figuring out the right search terms to drill down further.  Commercial sites like Wikinvest.com that use a similar approach can also be useful.
• The company site: If you’re researching a company, the company site is a good place to understand product lines and business units.  Many companies also have useful documents like case studies on their sites.
• Financial filings: The granddaddy of Securities and Exchange Commission filings, which all public companies in the US are required to file, is the 10-K annual report.  Companies include high-level descriptions of their businesses, competition, customers, and performance, although some go into more detail that others.  Another great resource for getting your bearings and understanding the industry.
• Investor presentations: These conference calls used to be limited to stock analysts, but in recent years regulation has forced companies to make them available to everyone.  These presentations can be great complements to financial filings, providing more in-depth information on current strategic initiatives and plans for the future.  Many companies now post presentations and audio files from the conference calls on the investor relations pages of their sites.
• Google Finance: This site provides a great, quick overview of the company, competitors, and financials for anyone strapped for time.  Similar sites like Tracked.com are also trying to take this aggregated information to a new level.
• Trade magazines: Finding the right trade magazine can be hit and miss since it can be difficult to tell which ones are authoritative versus just another website, and the authoritative ones sometimes charge exorbitant subscriptions if you want to access their articles.  Those issues aside, trade magazines are a great way to understand an industry, whether looking for recent competitive moves, company profiles, or industry-wide data.

Once you’ve perused these sources, you should have a better idea of how to organize your research when you do hit the search engines.  With more specific search terms, you can often avoid much of the irrelevant content that very general queries seem to attract.  In my next post, I’ll discuss organizing your research in a bit more detail.

What do you find to be the best place to start in your research projects?

Of course, there are plenty of tools out there that aim to cure these headaches.  Tools like Google are amazing, but they cause problems of their own.  We’ve all become habituated to going straight to search engines for answers, although they’re often not a good first resort.  More generally, the seemingly limitless availability of information has lead us to spend too much time searching for those elusive facts that will make or break our work.  Often, that time could be better spent analyzing the information and communicating it more clearly.  In the past, scarcity of information at least constrained business analysis to be simple and sometimes even elegant.  Now that constraint must be replaced by discipline on the part of the analyst.  Finally, search engines are in a constant race to maintain relevance, and often a variety of less relevant or redundant results end up ranking highly.

Since I’ve spent so much time and effort learning the ins and outs of business research, I’ll be sharing some of my experience and perspectives on how to do it effectively on this blog.  I’m also hoping to get other people’s perspectives in the comments.  I’ll also be examining various information sources, research tools, and best practices that are useful.  I’ll touch on some related topics like structuring business analysis appropriately.  Finally, since I’m working on developing better research tools for business users, I’ll be making announcements on new features as they come out.

If this sounds relevant to you, I’d be interested in your thoughts.  What research-related topics do you want to read about?