Archive for May, 2010
by Bill Ives
May 26, 2010 at 3:41 am · Filed under
Web 2.0
I recently attended the MIT Sloan CIO Symposium for the second time. It is an annual one-day conference, held on the MIT campus. I attended a session, Positioning IT as an Innovation Engine, led by Martin Reeves, Senior Partner, Boston Consulting Group. Panel members include: Ananth Krishman, CTO, TCS, Daxi Li, Chairman Chinese Association for Science and Business, Roy Rosin, VP of Innovation, Intuit, and Marco Orellana, CIO Codelco.
Martin said that this panel operates on the assumption that IT can drive innovation. This is still an issue at many firms. So he asked the panel if IT does drive innovation. Roy said that the essence of innovation today is to fail fast before investing a lot in the wrong direction. The CIO can be a big part of this. Get things done quickly to fail fast. The challenge is how to get into the right data to test ideas, how to set up a Web site quickly or do a prototype. Intuit uses IT to both create innovation and to run the company. For example, need to understand customer problems on such issues as how get billing processed faster. IT can help here.
Daxi said IT can drive innovation. CIOs play an important role in Chinese companies in innovation. For example, one Chinese company, a consortium from 19 universities, is expanding into new markets. IT can help collaboration between the parties and make processes more efficient.
Ananth discussed the gap between the potential of IT to drive innovation and what actually happens. Companies expect IT to drive innovation while making things run smoothly. If you do not do this part, you are not credible in the innovation space. Expectations are also around proving the platform for innovation and bringing things from the latest technologies. The credibility to introduce new technologies in dependent on success in the first two expectations. In his firm, they have separated the functions so the CIO runs things and the CTO does the other two, more innovation focused functions.
Martin raised the information overload issue. How do you shift through all the ideas that are generated? Ananth said he used to have to sort through the 30,000 new ideas that were suggested by employees. Now there is more control over this process but he still wants to capture the wisdom of crowds.
Roy said that innovation requires collaboration and accessibility. He looked at existing tools but found them to create silos. So they built their own tool, Brainstorm. I have reviewed it (see Intuit Brainstorm Offers Innovation Management). It is built around the notion of subscriptions so you see only the areas you want. They also look at the energy as many people circle around a topic. This is a signal to pay attention.
Martin said after awareness comes experimentation. For example, he said that some people will buy a Google ad word just to see if there is interest. In another example, their Turbo Tax people found that about 25% of people who get tax refunds do not have a bank account to receive it electronically, So they loaded the cash into a debit card and did hundreds of thousands of these rapidly through technology to meet this customer demand.
Daxi gave an example of a Chinese wind power company that wants to tap into wind at 10,000 meters high where it is much stronger. It could provide much of the world’s power needs. But how to access this power? It will be very expensive to test. IT allows for simulation to test ideas at a much lower cost.
Roy said that much of the innovation is a change management issue. Metrics come into play here, both project specific metrics and overall metrics. You need to pick what to measure, what to make visible, what to celebrate. This can change the culture of the company.
Ananth said there are metrics for incremental improvements such as ROI. Then there is innovation that is transformational and affects the entire company. These need program style measures such as time to market. For the truly disruptive innovations he measures things like failure rate and expenses. What is the throughput of new ideas? Also how does do the ideas affect the overall market?
Martin asked what culture needs to be in place for IT driven innovation? Daxi said that in China the CIO role is like an IT manager. He said that Chinese companies need to be able to establish a culture that encourages risk. Roy said it is about how to take intelligent risks and celebrating things that effect leading indicators and not simply rewarding large impacts on current revenue.
One questioner noted that only one member of the panel was actually a CIO. He asked if you need to set up a separate function to support innovation while IT is running the business. This was the case for two of the panel members. This seems to be an issue in many companies from what I heard today.
Roy pointed out that there are many types of innovation: product, marketing, commercialization, process. IT plays a different role in each. He pointed about the need to create slack in processes and mentioned an example using Quickbase to create process innovation for P&G. I reviewed another Quickbase example for XM Radio (see Changing Organization Behavior at XM Radio through Enterprise 2.0 and QuickBase). Putting transparency into the process was the big innovation here and that was enabled by an enterprise 2.0 technology. I think that using enterprise 2.0 tools is a way the companies can drive process innovation. Firms were IT plays a leading role in adopting these enterprise 2.0 tools provide a chance for IT to drive innovation if it focuses on this opportunity. If IT supports ways for the firm to capture process innovations that emerge and makes them easier to share and adopt, it can drive this type of innovation.
The audience said that money was less of a barrier to innovation than culture. Another big barrier was an obsession with efficiency so lessons are not learned.
by Bill Ives
May 18, 2010 at 3:41 am · Filed under
Web 2.0
Here is the second in a series of my notes on the 2010 Enterprise Search Summit. I attended a three part session. Leveraging Your Most Important Asset: Knowledge by Stacy Monarko, Sharpening Enterprise Search Performance, by Keith DeWeese, Taxonomy and Automated Indexing Project Manager, Chicago Tribune and Search Behavior Analysis by Harald Kirsh, Raytion. Here are the session overviews. My notes follow.
Leveraging Your Most Important Asset: Knowledge – ” We’ve heard it said, “An organization’s most valuable asset is its people.” Customer-driven firms will answer, “Customers.” Research-heavy organizations may answer, “IP.” The common element is “knowledge,” which comes from many places including customer interactions, employees, consultants, and reports. Learn what issues prevent most companies from exposing and utilizing information’s true value. Several case studies will highlight how significant, measurable value was gained with access to knowledge across global siloed information repositories.
Sharpening Enterprise Search Performance – “Current information volumes demand automated surfacing of relevant content and when such content is dynamic, intelligence must be embedded in the operations. Hear how the Chicago Tribune has successfully met this challenge: improving search engine effectiveness by automating key-wording and meta-tagging information, from web channels that cross 20 major US markets.”
Search Behavior Analysis – “While classical site analytics explores how visitors navigate to and within a website, search behavior analysis focuses on insights into what users are looking for in the first place. In addition to giving valuable hints on how to make search itself more effective, it can reveal gaps in the content and point to usability issues. This session will tell you how to implement search analytics in practice and how to integrate it with existing search engines.”
Stacy Monarko from Vivisimo discussed leveraging your most important asset: knowledge . It is more than search but information optimization. I would agree. Search is the means, a Bob B said in the last session, you want to find stuff, not search. To focus on just search is like Polaroid not wanting do digital photos as it would cut into their film business.
Instead looking at a search problem, Stacy said to look at a business problem like reducing the amount of redundant work by researchers. This is an information optimization question. Another business problem: How can I reduce training costs? How can I get people productive quicker? How can employees understand how to educate themselves. One information optimization project reduced training costs by 40% through addressing this question.
She discussed the need to sit in the user’s shoes. I found this to be the most fun part of knowledge related projects. Riding with English plumbers in the field was one great example for me.
Next Keith DeWeese from the Tribune Company discussed ontology and search at his firm. I was especially interested in what newspapers are doing on that they are under siege. Keith said his company has been operating under bankruptcy for the past few years. They have been using SAS tools since 2007. Content in news organizations is constantly changing.
They also have the challenge of helping others finding their content through search optimization. I was at a session by the Huffington Post where they are constantly running experiments in this area and rearranging headlines on an every 15 minute basis. That is hard for a traditional news organization to compete with. Keith mentioned the need to shift a headline from a “plane crash” to an “heroic landing” as the story evolves. There is also the issue of matching appropriate ads to content. They do not want dog food ads connected a story about a politician’s pet projects. They also offer guided search that provides customized classification for filtering the news to better match the user’s needs.
Next Harald Kirsh from Raytion in Germany discussed search behavior analysis without benefit of his slides. I felt for him as this happen to me last Fall in Montreal when there was also a screen resolution issue. I just ad libbed without the slides but he persisted and with help go the slides working.
Search logs help you see what people are looking for, not simply what they find. They selected some metrics to help with their search behavior analysis such as queries over time, most frequent queries, zero result queries, and most clicked results pages. It is hard to define what is a successful search as you do not know what the user is trying to do simply from looking at actions.
by Bill Ives
May 14, 2010 at 3:21 am · Filed under
Web 2.0
I recently attended the 2010 Enterprise Search Summit. This session, BI in the Age of Social Media, was lead by David Bean Ph.D., CTO, Founder, Attensity. Here is the session overview. My notes follow.
“Customers are flocking to the web to discuss everything from customer service complaints to product recommendations. JetBlue executives are mining unstructured text for insight into customer thoughts and actions. This session will discuss how companies can develop an efficient knowledge repository to enable more effective creation, maintenance, and administration of business knowledge from the web and other sources, and link it with all service-relevant documents. Presenters will provide examples of what worked and, more importantly, what hasn’t, in analyzing social media content.”
David changed his title at the start to BI in the Age of Social Media and How Search Can Help. Up until now search and BI have not been well connected. Then he said he has more of a liberal arts background and is not a tech person. I like this, as I am the same. He also said that h teaches linguistics but his degree not there. He also is a welder, raises chickens and fixes old tractors. I like him more already.
In the past search was simply put on the top of the BI tools to find reports generated by the tool. But it ignored the data behind the report.
Now social media has started to get into the BI space. This has happen largely with social media monitoring tools rather than traditional BI tools. I have looked at a number of these tools on the AppGap and agree with David.
David then asked why should BI care about search. First BI is about measurement. It quantifies business processes but has looked mostly inside. He said that now BI need to focus more on the Web as so many brand related conversations occur on the Web. There are “banking company x sucks” sites where people put their comments. People also do more research on what they plan to buy and they look at these comments by others.
There is now confessional intimacy. You look more at individual people who have just used the product than the professional rating systems. His firm looks at “first person intelligence” (their TMed term), what individual people say.
He gave a call center example to look at varying rates of call volume by region to discover why. The structured data is not enough to fully answer the question. You need to look at the call notes to see why the customers called. He wants to ask questions of both structured and unstructured data to go beyond the “what” to the “why.”
In addition to data like call center notes, you need to go out on the Web to forums, Twitter, Facebook, and other sites such as Craig’s List and YouTube. The Web volume is dramatically bigger for the web. They need a listening post to monitor this content and then the ability to integrate this Web data with a BI tool.
His firm bought a Web search tool, Biz360, to better look at this Web content. It is now Attensity 360. But there is need to shift through this volume as much is not relevant but some is. They are monitoring 37 million blogs as well as forums. They push all this data into a search engine. They vary case sensitivity in three ways. They have 3 billion documents hosted on Amazon and the index is updated every 30 minutes through Lucene. They focus on 3 – 6 months of the most recent data on a rolling basis.
They make 30,000 queries on this data every day. The queries can be up to 10,000 characters. The data is also pushed to their A5 analytic application. They use the queries to filter down what is important to their customers.
David discussed what has not worked – search as collector. They did a key word search on the customer’s products but it did not work as they got lots of porn and fetish sites when looking for Old Spice aftershave.
There is also the new language of social media and the abbreviations. They can do a table for this but there are other issues such as short hand like +1 to say I agree with the last comment in a forum. The traditional NLP problems of metaphor, sarcasm, and idioms now appear a lot of social media to make them harder. The same goes for typos such as missing spaces.
They have a team of linguists trying to keep up with the changes in languages, Darwin bypasses this issue, as it does not try to understand the content but lets the content self-organize and the user then can make the decisions. The translation and understanding issue is passed to the user but in a format that supports the user making these decisions. Now this is a different approach than looking at massive amounts of data in an automated manner so Darwin supports a different method of use. Each have their place.
Eric Andersen of IBM asked why they index everything instead of pulling out what the customer wants. David said they do not know what the customer wants in advance. New questions are always are emerging. So they look at the same data for all customers and just ask different questions. They create special queries for each customer, starting from standard queries and customizing them. Some of these queries are complex. Eric, who sat next to me, also told me about Emoticons, a new IBM tool for sentiment analysis.
by Bill Ives
May 12, 2010 at 3:20 am · Filed under
Web 2.0
Yammer was one of the first enterprise microblogging tools, and it continues to add features to extend the original concept of Twitter into an enterprise business application. A while back, I wrote about one case example, Implementing Enterprise Micro-messaging with Yammer at Océ., Océ is still an active user of Yammer. I was pleased to recently speak with Yammer’s VP of Marketing, Steve Apfelberg, about their capabilities and latest developments. Steve said that many of their new features came from customer input. You can see a sample screen below.
We began with an overview of some of the ways Yammer has extended the functionality of microblogging beyond Twitter to better serve business needs. First, they have not placed a 140 character limit on updates, or Yams, as they call them. There is a suggestion to keep things crisp but many business messages require more than 140 characters. I agree and have reluctantly avoided Twitter at times for this reason when it would have served a good purpose.
Next, Yammer allows for file attachments. Twitter allows for links to documents but often you want to use microblogging as an internal distribution channel for things like the latest sales presentation. These attachments are stored in the Yammer cloud and become searchable in their own right, a useful feature.
Steve said that Yammer allows for public or private groups within each Yammer network. This is not like Tweetdeck groups as the feature allows you to indicate who can receive what updates and provides different access levels. Yammer also has more social networking functionality than Twitter as there is an enriched profile feature. Samuel Dreissen at Océ said that the Yammer profiles helped spread usage. Samuel said he got more compliance with the profile completion in Yammer than in knowledge management systems because of the simplicity of the form. As new users complete their profiles, the system guides you to invite your colleagues, which support the viral adoption of Yammer.
To further enhance social networking, Yammer also offers org charts that connect profiles. You can see how participants are connected and this can help you decide who to follow. Yammer provides suggestions on who you might want to follow that are partially based on the org charts. You can see a sample org chart below.
Yammer provides auto updates or Yams based on user actions such as joining a group or adjusting an org chart. There is two-way email integration. Steve said they did this since some people still live within email. You get email notifications if you want and can post to Yammer from email.
Users can “like” an update. This will then display the update in the message feed of all of their followers who do not follow the original message poster. I really liked this feature as it avoids spamming people with redundant messages but also does the work for you of trying to figure out who already saw the message and who did not.
You can use tagging through a hashtag in the same way you do with Twitter. Steve said this is something frequently used within their own firm to segment information into searchable topics.
We next discussed their latest major feature, Yammer Communities. Steve said that many customers who were using Yammer for internal communication asked them if they could additionally utilize Yammer to collaborate with their external business network, including customers, partners, suppliers, and vendors. Until now, this was not possible as Yammer required all members of a Yammer network to share a common, work email domain.
Now Yammer Communities enables companies to create a new type of Yammer network that is not restricted to a common email domain. The Communities option provides companies with a secure, private, and separate space to communicate with their external business contacts. Membership, messages and other data in the Community are completely separate from membership, messages and data in the parent company network; users must toggle between networks to access these separate Community networks. It has all the features of the traditional or “canonical” instance of Yammer. People within a company can use both. Steve gave the example of the Dana Farber Cancer Institute in Boston that has used Yammer Communities to allow newly diagnosed cancer patients to communicate with cancer survivors in a secure manner.
Yammer provides both a free version and premium paid versions with expanded features. More than 10,000 companies and organizations created new Yammer networks in Q1 2010, bringing the total to more than 70,000. More than 70% of Fortune 500 companies are using Yammer and there are over 800,000 users. This is a testimony to the power of microblogging and to what Yammer brings to this capability.
by Bill Ives
May 6, 2010 at 3:23 am · Filed under
Web 2.0
I have covered NewsGator here before (see NewsGator Offers Social Sites ™: A Comprehensive Enterprise Social Networking Suite). I have been a fan of NewsGator for some time as they were my first RSS reader in early 2004 (see Finally, I’m Scanning the Web through NewsGator). I recently spoke with Brian Kellner, VP of Product Management to catch up on their latest moves.
We started with the iPhone offering. NewsGator is offering a native social iPhone client for Social Sites on SharePoint. They have wisely taken a subset of the Social Sites features, picking the ones most likely to be used through a smart phone. The features center around a sophisticated microblogging suite to provide a focused activity stream that goes across applications. The idea is to provide a central place to receive updates of all types, comment, and pass them on.
Brian recently wrote about this concept in a blog post, “The Trojan Horse Battle for Your Event Store.” He writes that, “in only a few years, a central event store will be a standard part of nearly all enterprises, and that architectural decision currently is being made indirectly at many organizations.” I certainly agree with this and have written related posts on several occasions. He goes on to write that, “microblogging seems to be achieving its greatest value in the context of broader activity streams that serve as a kind of nervous system for the company.” This broader role for microblogging, as it becomes more widespread within enterprises, represents a great opportunity for both users and vendors. NewsGator has recognized this and is offering an application to provide the unifying activity stream or “event store.”
The latest NewsGator application for the iPhone is a large step in this direction. You can attach updates to an object or a link. You can attach an update to a context such as an event or a project, and you can react to updates from others, maintaining the context. To support these three capabilities, mobile users can use an iPhone to send status updates, videos and photos to Social Sites for distribution. They can comment on and “like” updates they value and see notifications when others comment on or like their updates. Below you can see an example of the iPhone interface.
Participation in community discussions is also enabled and users can target a question to a person or group. They see the updates that are radiating to others, view updates from a specific colleague or community, and look up contact information for colleagues. Below you can see comments on an activity. They can begin threaded discussions to maintain the context.
They are now planning a BlackBerry version to extend their smart phone coverage. The initial iPad client will contain all the iPhone features. NewsGator is also exploring what additional capabilities, such as content creation, should be added to the iPad client.
The enhanced microblogging features are also part of the desktop version of Social Sites 3.1 that is fully integrated with SharePoint. This SharePoint integration provides enhanced security and leverages existing SharePoint features such as the user profile, reducing redundant requirements. Below is an example of a Social Sites workspace within SharePoint. The iPhone version provides the functionality shown in the central column, while the web version offers the broader content in the three columns.
You can also have Social Sites as a desktop client independent of SharePoint and within your desktop as shown below. In addition, you can work with Social Sites through email notifications and replies. You can also get activity digests through email and respond back still within email.
We next turned to the enhanced analytics that Social Sites now offers. The activity stream sets up a rich opportunity for analytics around the conversations and knowledge work within the enterprise. I think this is one of the major benefits of operating within an Enterprise 2.0 environment. NewsGator has recognized this and has done some nice work with data visualization. For example, you can get a tag cloud (as shown below) of the top 100 themes within the activity stream.
Clicking on any of these terms brings you to the instances of their use. You can also see the top ten experts on each term (as shown below). The person closest to the term in the middle is the most active. Several factors such as content authored, ideas generated, and tag usage weighs into the expertise scoring. You can see specific contributions for each expert in the right column.
This allows you to quickly see who is saying what to whom and stay better attuned to the pulse of the organization. There are a number of other creative visualizations. For example, you can drill down into the actions of an individual in the chart above.
The enhancements to NewsGator demonstrate some of the exciting business possibilities that are arising as the consumer Web 2.0 tools get greater usage within the enterprise. I look forward to seeing more.
by Bill Ives
May 3, 2010 at 3:29 am · Filed under
Web 2.0
Before there was enterprise 2.0 I used to think that, except for utilities like email, most business information-based software such as knowledge management needed to be process aligned. I have softened that position somewhat but for certain tasks, process-aligned remains the key to effectiveness. This is the approach taken by the Clearwell E-Discovery Platform, a product targeted at information-related legal process such as discovery. I recently spoke to Kamal Shah, the Clearwell VP of Products and Marketing, on how this product provides end-to-end process support for – early case analysis, search, cull, review, redaction and production.
Clearwell was founded in 2004 and they are about to release version 5.5 of their eDiscovery product. They now have over 500 customers including Cisco, Constellation Energy, Microsoft, BP and NBC, and they work with 42 business partners. They offer the software directly to enterprises and also sell through business partners who provide Clearwell as a hosted service. Clearwell provides an appliance based solution that allows for rapid deployment, often in about 30 minutes.
Kamal explained that the industry has adopted an Electronic Discovery Reference Model to promote a common vocabulary and other standards within the space. Clearwell supports this model and provides a common platform for the various stakeholders in the eDiscovery process including legal, IT, HR, audit, security, and compliance.
Kamal went on to describe the eDiscovery process and how Clearwell supports it. First, there is the initial data collection. Once data is collected, then Clearwell provides pre-processing capabilities. At this stage you also want to understand the task at hand and effectively estimate the cost and duration. Clearwell is able to go after multiple data types and sources to ensure effective retrieval.
Figure 1: Timeline Analytics Quickly Identify Gaps in Collection Across All Data Sources
Next comes the processing. Clearwell has ramped up their processing speed to help with this effort. With the new 5.5 Release, Clearwell can process data at speeds of over 1 terabyte per day and scales to 100 million documents on a single appliance. De-duplication also occurs in this step as you want to reduce the amount of information that your legal counsel, operating at high hourly rates, has to sift through. Part of this is eliminating copies of redundant content such as duplicate files and multiple copies of email forwarded around. At the same time you want to know who knew what and when in these email chains. Clearwell sets them up as a threaded discussion to easily uncover the answers to these questions.
Figure 2: Discussion Threads Identify All Participants and Who Knew What and When
Now comes search and cull down to further refine the content. Usually only 10 to 20% of the content is relevant so you want to get there as efficiently as possible. You can set up auto-filters to eliminate known false positives such as an article on the topic that was forwarded around but not relevant to the case. Next step is search. A firm might get several hundred keywords from a government agency such as the SEC or from an opposing litigator. The trick is to provide all of the relevant information and no more. If you provide too much you may be alerting the opposition to potentially damaging information not relevant to the inquiries. Providing too little may result in damaging later discoveries and penalties for failing to properly comply. Clearwell provides robust keyword and concept search capabilities that help you fulfill the request quickly and easily.
Figure 3: Transparent Search Reduces False Positives of Keyword Search Results
After distilling down the content, Clearwell supports the legal review process with nested folders, decision tree tagging, and indicated related documents. There is a simplified review interface to speed the process. Finally, the post-review production phase is supported through such features as integrated production folders, image based production, custom labeling using headers, footers, and watermarks.
Figure 4: Review Capabilities Allow Users to Quickly Redact Privileged or Confidential Data
Kamal said that the cumulative effect of these process-aligned supports can result in significant costs savings.
Because of the ease of implementation, Clearwell lets potential clients set up test cases to determine their own ROI and this usually results in a new client. I can see why. It certainly makes to sense to use a product like Clearwell in your enterprise legal processes. This is one situation where process aligned remains a requirement for success.
