The future of enterprise search has been referred to as search and business intelligence (BI) for a couple of years. As such, the future of enterprise search isn't forecast to be a radical set of new search features but the reconciliation of complementary silos. Of course, we never know the future it gets here, so now that early adopters are a few years into the search plus BI vision, we can begin to see what it means. However, dialing back in time, three possible outcomes for search plus BI were predicted:

 

Frankenstein: Bolt together the brain of one into the body of another and give birth to a monster. In search and BI terms, a big BI company buys a small search company and adds a search box to its reporting engine or vice versa. It fulfills the request for proposal, but the shallow integration does little else.

 

Surf 'n Turf: Lobster is delicious, and filet mignon is delicious, and so the two together must be more delicious - yet the sum turns out to be less than its parts. In search and BI terms, advances in technology let search vendors chart and visualize textual results or let BI companies analyze unstructured documents. The integration works, but the features - novel as they may be - don't solve a pressing business challenge. Thus the combined technologies become bloated.

 

Steel: Steel made it possible to construct much stronger buildings than with brick making it possible to build taller by orders of magnitude. That technological change opened up green-field possibilities - no less than the rise of modern cities. In search and BI terms, there is a new way to solve information intensive problems that had been intractable, and that in turn makes possible new business processes that give early adopters a competitive advantage worth billions of dollars. 

 

Which One?

 

So far, all three types of search plus BI can be found in the wild. Early adopters ferreted out Frankenstein and surf 'n turf from steel, and now the next wave of adopters benefit from a sharper view of the green field in front of us. So what does it look like?

 

To go back to our building analogy, the first steel-framed buildings had brick walls, which added needless weight and expense because brick was no longer needed to bear any load. It took a few years to realize that much cheaper, lighter walls like floor-to-ceiling windows were the future. Likewise, the first step to seeing the Search + BI vision is to drop the constraints of legacy technologies.

 

Search and BI live in separate silos because when their respective foundations were born in the 1970s - the inverted index and relational database, respectively - they needed to treat unstructured and structured data differently to meet the price/computing constraints of the day. Over time, these became more than just application silos. They became business process silos with separate standards, knowledge bases, application integrations, vendor relationships and many other nontechnical artifacts. As a result, those early architectural commitments have left us with constraints that are no longer simply technological.

 

Three main things have changed since those early architectural commitments created search and BI silos.

 

People: People expect the simplicity of Web applications during work hours, too. They don't expect a training course or even a help file - they expect to intuitively click around in a zero-training application.

 

Data: To get 360-degree visibility into all the information a worker needs to make a decision, he needs access to many source systems and formats. But forcing data into structured or unstructured silos suits application constraints, not business needs. In reality, business data is semistructured, a hybrid of relational data from multiple sources and schemas, documents with fielded information, unstructured documents that have been augmented with structure and, increasingly, piles of XML - all spread across the enterprise. Information visibility should account for the reality of today's data.

 

Computing: The current price-computing curve is a million times more powerful than when search and BI were first architected. While search and BI have improved incrementally since then, early architectural commitments make them unable to take full advantage of these hardware leaps. Cheap memory and 64-bit multicore processors are putting supercomputers into the range of commodity prices. Only a fresh architecture can harness that new power.

 

The "steel" version of search plus BI is an architecture designed around the Web expectations of today's users, the semistructured data in today's enterprise and the new supercomputers in the data center. With that new architecture, we can finally offer workers complete visibility into enterprise data of all kinds, allowing them to make informed decisions where they had been making gut decisions before.

 

So What?

 

Search plus BI isn't about faster search or higher-scale BI. Like steel enabling cities, "steel" search plus BI is making possible entirely new business processes. And with those, early adopters have been reporting results measured in the hundreds of millions of dollars and beyond.

 

The future of search plus BI is in the competitive advantage it offers early adopters that are envisioning business processes that were never before possible. By taking a fresh architecture and giving everyday users unprecedented visibility into today's enterprise information, we're seeing the next great source of efficiency that IT can bring to the business.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access