Remember The Jetsons? The flying cars and the automated kitchen and the food pills? Sometimes modern life can feel like that futuristic utopia. We've got robots in the home and a speech-recognition personal assistant named Siri built in the iPhone in our bag. IBM Watson, a supercomputer, beat its human competition in the TV game show Jeopardy! last year. How? By translating corny, nuanced questions into a format it could understand and compute.
But for most of us, our digital experiences at work feel like we're stuck in The Flintstones.
We wonder: "How can Amazon.com monitor my customer data so closely that it knows what book I want next, but after five years of daily use, my enterprise search engine doesn't get that I work in HR in the Chicago office?" We need to dig into our enterprise information so it is more rich and useful. Hal Varian, Google's Chief Economist, explained in the McKinsey Quarterly that "We have free and ubiquitous data, so the complementary scarce factor is the ability to understand that data and extract value from it." (He even goes so far as to say that statisticians will be the sexy job in the next 10 years!)
It's understandable to be cynical about semantic processing, especially if you've been told it relies on manually entered metadata.
In 2007 Larry Elison said: "We think the paradigm for doing business, how people do their daily jobs is changing and is moving to a search paradigm.” For years Oracle has worked on weaving its search functionality into and across Oracle applications. It's called Secure Enterprise Search (SES) and it's invisible to Content & Collaboration (C&C) professionals because it's inside the Fusion platform, rarely sold as a standalone solution. With SES integrated in Oracle products, Oracle envisions "action-oriented" enterprise search. What does that look like? When workers don't just search for pending expense reports, they also can pay them from the search UI.
When search is an embeddable service, it makes it easier to use search to get tasks done. This is why I think infrastructure vendors (HP, Oracle, Microsoft, Dassault) acquiring specialized vendors (Autonomy, Endeca, Fast, and Exalead, respectively) is a good thing for C&C professionals. What's missing from these marriages? Semantic search capabilities -- where search surfaces unstated concepts and allows users to visualize the patterns and trends locked inside volumes of text. (IBM is one to watch for this vision -- a leader in BI, they have recently commingled their search and content analytics technology to create a new product.)
Perhaps no one understands better than Dan Ranta, Director of Knowledge Sharing at ConocoPhillips, that the challenge of sharing knowledge is very real — while the potential payoff can be large. Seven years ago, ConocoPhillips launched a large initiative to create internal communities of practice that would enhance knowledge sharing within the firm. With operations in more than 30 countries, encompassing job sites often in remote locations, the international energy company knew that to continue on its success trajectory, it needed to rapidly and effectively harness the knowledge of its highly skilled but geographically distributed workforce.
Today, the ConocoPhillips' knowledge-sharing program — built upon 150 global "networks of excellence" — is ranked as best-in-class across industries, and has documented hundreds of millions of dollars in estimated cash flow from its start in 2004 to the present. To learn more about how firms can drive business excellence with formal, global networks, I spoke with Dan in preparation for his keynote this week at Forrester’s Content & Collaboration Forum.
1) Can you explain the reasoning behind the proactive and reactive components of your networks of excellence?
What's a customer-obsessed company? One that is deeply committed to know and engage with its customers. The three winners of our 2011 Voice of the Customer award -- Adobe, Fidelity and JetBlue -- don't just train employees to deliver great customer experiences; they monitor service satisfaction and systematically act on what they learn. My colleague Zach Hofer-Shall calls this management and analysis of customer-generated information "Social Intelligence."
I think the Voice of the Employee should share the spotlight with the Voice of the Customer.
Few clients I talk to analyze employee-generated information the way that they do customer-generated information. It's now mainstream to listen to customer opinions regarding your product's or service's shortfalls or what competitors do better. But it's cutting-edge to listen to employees as part of a consistent, automated, scalable, strategic initiative. I am not talking about reading private emails or sending an annual employee survey. Instead I mean mining solicited sources like open-ended feedback requests and unsolicited sources like wikis, content archives and public internal social profile pages.
Yesterday, HP agreed to buy UK software firm Autonomy Corp. for $10 billion to move into the enterprise information management (EIM) software business. HP wants to add IP to its portfolio, build next-generation information platforms, and create a vehicle for services. It is following IBM’s strategy of acquiring software to sell to accompany its hardware and services. With Autonomy under its wing, HP plans to help enterprises with a big, complicated problem – how to manage unstructured information for competitive advantage. Here’s the wrinkle – Autonomy hasn’t solved that problem. In fact, it’s not a pure technology problem because content is so different than data. It’s a people, process problem, too.
Here is the Autonomy overview that HP gave investors yesterday:
Of course, this diagram doesn’t look like the heterogeneous environment of a typical multinational enterprise. Autonomy has acquired many companies to fill in the boxes here, but the reality is that companies have products from a smorgasbord of content management vendors but no incentive to stick with any one of them.
He highlights text analytics technology in the report because understanding unstructured data plays a critical part in daily operations. Enterprises have too much content to review and annotate manually. Text analytics products from vendors like Temis and SAS mine, interpret, and add structure to information to reveal hidden patterns and relationships. In my 2009 overview of text analytics, I cite the primary use cases for these tools: voice of the customer, competitive intelligence, operations improvements, and compliance and law enforcement.
But there are a few other sweet spots for text analytics tools in the enterprise:
Analytics and search: Analytics tools surface and visualize patterns; search tools return discrete results to match an expressed need. But these disciplines are blending. People want to drill in to high-level analysis to find the specific thing customers buzz about. And many searchers don’t know how to articulate their need as a query and are looking for the big picture on a topic or trend. Forrester expects these solutions to come together, as search tools mainstream semantic features like entity extraction out of the box, and analytics vendors introduce new ways to investigate relationships and data output.
Gene briefly explores the misunderstanding between “Enterprise IA” and “User Experience IA.” This tension was well characterized by Peter Morville almost 10 years ago (See “Big Architect, Little Architect.” Personally I think it’s clear that content is always in motion, and unsupported efforts to dominate and control it are doomed. People are a critical element of a successful IA project, since those who create and use information are in the best position to judge and improve its quality. Many hands make light work, as the saying goes.
For example, if you want a rich interactive search results page, you need to add some structure to your content. This can happen anytime from before the content is created (using pre-defined templates) to when it is presented to a user on the search results page. Content is different than data, a theme Rob Karel and I explored in our research on Data and Content Classification. For this reason, IA is both a “Back end” and a “Front end” initiative.
The technical folks behind Monster.com invited me to visit last week. I somehow couldn’t convince them to show me any Superbowl ads but they did demo their cool new search engine. It’s based on technology they acquired when they bought Trovix in 2008. What can it do?
Understand the meaning of words: The search engine knows the difference between “development” in the fundraising context and “development” in the software context.
Appreciate the relationships between words: A custom ontology fortifies the search engine. The ontology rolls up skills like auditing into the larger category of finance. It differentiates between a top ranked school and a lower ranked school. It understands that years spent working as a prosecutor should count towards a candidate’s overall legal experience.
Cut text-heavy resumes into nimble content components: Recruiters can use the power resume search to compare candidates side-by-side, because the search mixes and normalizes the information into simple, clean categories like “Experience,” “Education,” and “Skills”.
Our latest featured podcast is Leslie Owens's "The Use Of Text Analytics To Mine Unstructured Content."
In this podcast, Leslie sheds light on the tools and resources available to analyze and classify “unstructured text,” such as emails or survey documents. These tools could yield solutions to business problems as an add-on for business intelligence tools, or for customer relationship management.
Big news in the information management world today – Autonomy announced it will acquire Interwoven for $775 million.
Since 2005, Autonomy has acquired technology for search (Verity), archiving (ZANTAZ), and records management (Meridio). With Interwoven, Autonomy gains a technology foothold where it was previously weakest -- at the point where digital content gets created, captured, and managed. Yet knowing Autonomy, it’s likely after Interwoven’s solid customer base in several niche market segments: law firms and customer-facing media, entertainment, and commerce Web sites. All of these Interwoven customers had better prepare for a knock on the door from Autonomy reps prepared to sell them on the virtues of extracting “meaning” from their digital information (using Autonomy IDOL, of course).
Enterprise search and enterprise content management are two sides of a coin. Both are necessary to create, manage, store, find and analyze information. Yet information workers still generate an enormous amount of content in word processing applications and distribute it via email. Content created in this way is difficult to manage and control as well as difficult to find. The high price Microsoft paid for FAST Search and Transfer last year was based in part on the expected value of combining the two sides of the coin — to tightly integrate search and classification capabilities at the point where content is created and accessed. Autonomy brings more sophisticated — and much needed — archiving and records management capabilities to this picture.