The advantages of integrated search vs. federated search in market intelligence applications

Northern Light delivers integrated search, which is very different from the widely implemented federated search solutions available from many well-known vendors. That difference is fundamental to the value proposition of our strategic market research portals.

The biggest advantage of federated search is that its “lowest common denominator” approach is easy to implement and the technology is widely available. In a federated search, a search engine sends the user query to the search engines that are native to each source being queried. It receives back a search result from each source (in, say, an XML package), organizes the various search results together according to algorithms it imposes on the universe of results using such factors as “first responder wins” or words in the title and presents the results to the end user.

For many applications, there is nothing wrong with the lowest common denominator approach of federated search . But the demands of strategic market intelligence, market research, and business decision making require something more — search that can understand, analyze, and triage content to surface meaning and actionable intelligence. They require integrated search.

The power of integrated search begins with content aggregation

Search quality begins with the content searched. Because Northern Light knows the use cases of our clients before the first search is ever initiated, we tailor the content, search technology, and analytics to those uses.

We aggregate all the content by obtaining and indexing a full-text original copy of every document from every source, using perfectly consistent indexing, taxonomy, and text analytic strategies uniformly across all sources. The result is an integrated index that behaves the same way across all sources in terms of search syntax and relevance ranking. This level of consistent, rationalized indexing across otherwise unrelated sources enables compelling analytics and meaning extraction in a way that would be impossible across simple federated search.

Northern Light’s integrated approach

  • Indexing strategies are consistent across all sources.
  • Search syntax is consistent across all sources. All sources can be searched using the same advanced, flexible, and powerful search syntax. Alerts can also use this same consistent syntax and be depended on to work the same way for all sources.
  • Taxonomies can be applied across all sources. With Northern Light’s integrated search strategy, a client can use a relevant taxonomy applied to all sources in the research portal. Control of taxonomies resides entirely with the client, not with the content sources. This capability amplifies the search value in competitive intelligence analysis and market intelligence research.
  • Text analytics is enabled across all sources. Because the search engine has access to the full-text of the documents, Northern Light’s powerful MI Analyst text analytics can automatically extract meaning, surface concepts, identify relationships, and cut time to useful insight while improving decision quality. Northern Light’s MI Analyst does a full-text “read” of all the research reports and articles returned on a results list, identifies the concepts and issues that they report on, and directs the researcher to the most interesting documents — not from a search relevance perspective, but from a meaning perspective.
  • Relevance ranking is consistent across all sources. Unlike general purpose enterprise search engines that must serve many dissimilar use cases, we have the luxury of knowing that our users will be professionals doing business, technology, and industry-specific research on large repositories of market research reports and news stories. This foreknowledge allows Northern Light to optimize our relevance ranking for these purposes. Northern Light’s relevance ranking balances business-critical factors, and can be customized to meet the needs of our clients.