Enterprise Search 2.0?

While visiting Enterprise Search Summit in San Jose I realized that enabling Enterprise 2.0 within enterprise search is the hottest trend at the moment. Is it Enterprise Search 2.0?

Andrew McAfee who coined the term Enterprise 2.0 and has released a book on the subject, spoke about how to use altruism to develop the enterprise. People are wired to help and if we stop obsessing about the risks and lower the bars for how people can help each other it is possible to make this work within a corporate environment.

He also spoke about how process control and how much workflow control. How much do we really need? Make it easy to correct mistake instead of making it hard to make them. With regards to innovation he pointed out that we need to question credentialism and build communities that people want to join. To leverage the intelligence aspects within the enterprise we should explore and experiment with collective intelligence such as prediction markets and open peer review processes. All in all make it easy for people to interconnect.

Very high improvement in access to knowledge, internal experts, satisfaction, increased innovation and customer satisfaction.

I also recommend to read Price Waterhouse Coopers Technology Forecast Summer 2008 to get a good overview of the available tools and technologies.

So how does this impact enterprise search? Search can be made to be the facilitator for Enterprise 2.0. Of course it is possible to index and make all blogs, wikipedias, tweets (yammer), online communities and social networks searchable, but that is only one way to make it this new environment more findable. If someone tweets or blogs about information we should use that information to impact on the search results and ranking. We could also track user behavior on a site to make certain information more visible with regards to implicitly expressed interests.

High Expectations to Googlify the Company = Findability Problem?

It is not a coincidence that the verb “to google” has been added to several renowned dictionaries, such as those from Oxford and Merriam-Webster. Search has been the de facto gateway to the Web for some years now. But when employees turn to Google on the Web to find information about the company they work for, your alarm bells should be ringing. Do you have a Findability problem within the firewall?

The Google Effect on User Expectations

“Give us something like Google or better.”

 

“Compared to Google, our Intranet search is almost unusable.”

 

“Most of the time it is easier to find enterprise information by using Google.”

The citations above come from a study Findwise conducted during 2008-2009 for a customer, who was on the verge of taking the first steps towards a real Enterprise Search application. The old Intranet search tool had become obsolete, providing access to a limited set of information sources only and ranking outdated information over the relevant documents that were in fact available. To put it short, search was causing frustration and lots of it.

However, the executives at this company were wise enough to act on the problem. The goal was set pretty high: Everybody should be able to find the corporate information they need faster and more accurately than before. To accomplish this, an extensive Enterprise Search project was launched.

This is where the contradiction comes into play. Today users are so accustomed to using search as the main gateway to the Web, that the look and feel of Google is often seen as equal to the type of information access solution you need behind the firewall as well. The reasons are obvious; on the Web, Google is fast and it is relevant. But can you—and more importantly should you—without question adopt a solution from the Web within the firewall as well?

Enterprise Search and Web Search are different

  1. Within the firewall, information is stored in various proprietary information systems, databases and applications, on various file shares, in a myriad of formats and with sophisticated security and version control issues to take into account. On the Web, what your web crawler can find is what it indexes.
  2. Within the firewall, you know every single logged in user, the main information access needs she has, the people she knows, the projects she is taking part in and the documents she has written. On the Web, you have less precise knowledge about the context the user is in.
  3. Within the firewall, you have less links and other clear inter-document dependencies that you can use for ranking search results. On the Web, everything is linked together providing an excellent starting point for algorithms such as Google’s PageRank.

Clearly, the settings differ as do user needs. Therefore, the internal search application will be different from a search service on the web; at least if you want it to really work as intended.

Start by Setting up a Findability Strategy

When you know where you are and where you want to be in terms of Findability—i.e. when you have a Findability strategy—you can design and implement your search solution using the search platform that best fits the needs of your company. It might well be Google’s Search Appliance. Just do not forget, the GSA is a totally different beast compared to the Google your users are accustomed to on the Web!

References

http://en.wikipedia.org/wiki/Googling

Designing a Good Search Experience – Summer Reading

The people at Findwise are entering vacation mode one after the other. While finishing up my projects before summer vacation I started thinking about what are the important parts of creating a good search experience. So I wanted to give you a few tips before leaving the office for the summer.

Myself and Caroline participated at Business to Buttons in Malmö in June. I met a lot of talented people and had lots of interesting conversations. One of the topics i ended up discussing the most was: Search is just search, right?

A very common opinion amongst designers is that search is just search. You put a search box in the upper right corner and then you’re done. The search engine has thought of everything else, hasn’t it? I found myself arguing about two things that are very close to my heart:

  • Choosing the righ search platform
  • Designing a good search experience

Choosing the right platform

There is a difference between search engine platforms. You just don’t go out and by one and think that’s it. “Search is fixed.” It does matter what platform you choose! Depending on your choice you can tune it in different ways to fit your needs. You don’t just install Google or any other platform for that matter, and think your done. If you do, you’re in trouble. As Caroline wrote about in a previous blogpost, most enterprise search projects with problems, have problems that are not related to the platform but to the fact that the organization does not have a strategic way of working with search.

To give you designers and other design interested people a quick start to this subject I recommend listening to a podcast from Adaptive Paths UX week 2007 where Chiara Fox talks about search and interaction design. (You can download the podcast from Itunes store for free.) It will introduce you to some of the basic things to think about when it comes to getting what you want from your search engine.

Designing a good search experience

When designing a good search experience there are lots of things you should think of. But without getting to involved in advanced filters, navigators, query suggestions and other things you first need to fix the basics. Showing relevant information in the search results. One of the most common problems I meet at new customers is search results lists that make it practically impossible for the users to understand what the result is without clicking on it. All search results look the same no matter if they’re documents, web pages, people, applications, or products. The only way for the user to understand what information they can find in the result is by clicking on it. A search application that forces the user to use pogosticking is in no way better than using poor navigation. So first you need to think about what information needs to be displayed about different types of search result. What information is relevant for a document, or for a web page?

To get you started thinking about this I recommend reading the articlefrom UIe about creating good search results. It will introduce you to some of the basics.The article describes web site search. Enterprise search is off course more complex since you have more types of sources but the basic idea is the same: Show the user the information they need.

So that was two recommendations for your reading list this summer (in case there is a rainy day or two).

If you have any question about choosing the right platform or design good search experiences please contact us. More on these topics will also come after the summer.

From the people here at Findwise, have a great vacation everyone!

What Differentiates a Good Search Engine from a Bad One?

That was one of the questions the UIE research group asked themselves when conducting a study of on-site search. One of the things they discovered was that the choice of search engine was not as important as the implementation. Most of the big search vendors were found in both the top sites and the bottom sites.

So even though the choice of vendor influences what functionality you can achieve and the control you have over your content there are other things that matter, maybe even more. Because the best search engine in the world will not work for you unless you configure it properly.

According to Jared Spool there are four kinds of search results:

  • ‘Match relevant results’ – returns the exact thing you were looking for.
  • ‘Zero results’ – no relevant results found.
  • ‘Related results’ – i.e. search for a sweater and also get results for a cardigan. (If you know that a cardigan is a type of sweater you are satisfied. Otherwise you just get frustrated and wonder why you got a result for a cardigan when you searched for a sweater).
  • ‘Wacko results – the results seem to have nothing in common with your query.

So what did the best sites do according to Jared Spool and his colleagues?
They returned match relevant results, and they did not return 0 results for searches.

So how do you achieve that then? We have previously written about the importance of content refinement and information quality. But what do you do when trying to achieve good search results with your search engine? And what if you do not have the time or knowledge to do a proper content tuning process?

Well, the search logs are a good way to start. Start looking at them to identify the 100 most common searches and the results they return. Are they match relevant results? It is also a good idea to look at the searches that return zero results and see if there is anything that can be done to improve those searches as well.

Jared Spool and his colleagues at UIE mostly talk about site search for e-commerce sites. For e-commerce sites bad search results mean loss of revenue while good search results hopefully give an increase in revenue (if other things such as check out do not fail). Working with intranet search the implications are a bit different.

With intranet search solutions the searches can be more complex when information not items, is what users are searching for. It might not be as easy to just add synonyms or group similar items to achieve better search results. I believe that in such a complex information universe, proper content tuning is the key to success. But looking at the search logs is a good way for you to start. And me and my colleagues here at Findwise can always help you how to get the most out of your search solution.

The Multitasking Man

The first Keynote on this years British HCI conference was by Stephen Payne who talked about task switching and studying of multiple texts. He talked about how users use different strategies to find the information they’re looking for and how it is important to give clues to which content is the most relevant.

So in this world that’s overloaded with information, how do you choose the best text?

According to Stephen Payne, most people use a satisficing strategy when reading trough texts. That basically means that you study a good enough text, instead of keep looking for the best one. So you pick the first text and keep reading until you’ve learned enough or get fed up. The consequences of this is that it’s important to create texts that have good readability and that are skimable as well, which can be made by dividing the text into smaller patches.

One other thing Stephen Payne talked about was the fact that the give-up time for a task is depending on the sucess with the information seeking in the previous results. So if the first page the user looks at is good, the user has more patience looking for information in the next page. And so if the first results are bad the user’s patience and give-up time decreases.

The consequences this has for search is related to content tuning and ranking. Because if the first results are bad the users have less patience looking for the information they really wanted. By providing sponsored links for example, one could control the top hits for specific keywords and then make sure that the most popular searches get good results. This does not have to be the way to go, but it is absolutely an option one could use to aid ranking of documents and improve the user’s perceived quality of the search results.

Find People with Spock

Today, Google is the main source for finding information on the web, regardless of the kind of information you’re looking for. Let it be company information, diseases, or to find people – Google is used for finding everything. While Google is doing a great job in finding relevant information, it can be good to explore alternatives that are concentrated upon a more specific target.

In the previous post, Karl blogged about alternatives to Google that provides a different user interface. Earlier, Caroline has enlightened us about search engines that leads to new ways on how to use search. Today I am going to continue on these tracks and tell you a bit about a new challenger, Spock, and my first impressions of using it.

Spock, relased last week in beta version, is a search engine for finding people. Interest in finding people, both celebreties and ordinary people has risen the past years; just look at the popularity of social networking sites such as LinkedIn and Facebook. By using a search engine dedicated to finding people, you get more relevancy in the hits and more information in each hit. Spock crawls the above mentioned sites, as well as a bunch of others to gather the information about people you want to find.

When you begin to use Spock, you instantly see the difference in search results compared to Google. Searching for “Java developer Seattle” in Spock returns a huge list of Java developers positioned in Seattle. With Google, you get a bunch of hiring applications. Searching for a famous person like Steve Jobs with Google, you find yourself with thousands of pages about the CEO of Apple. Using Spock, you will learn that there are a lot of other people around the world also named Steve Jobs. With each hit, you find more information such as pictures, related people, links to pages that the person is mentioned on, etc.

In true Web 2.0 fashion, Spock uses tags to place people into categories. By exploring these tags, you will find even more people that might be of interest. Users can even register on Spock to add and edit tags and information about people.

Over all, Spock seems like a great search engine to me. The fact that users can contribute to the content, a fact that has made Wikipedia to what it is today, combined with good relevancy and a clean interface it has a promising future. It also shows how it is possible to compete with Google and the other giants at the search market by focusing on a specific target and deliver an excellent search experience in that particular area.

Interesting New Search Features

Out on the web there are a large number of small search engines that try to stand out and maybe take some of the market shares from Google. Many of them have interesting search features.

I would like to introduce some of them in order to help other realize that search can (and should) be a bit more then a search bar and a list of hits. A number of these alternative search engines have focused on the visual presentation of the search result in interesting ways. For example the search engine quintura uses tag clouds of related terms and concepts to the original query.

A slightly different approach has been taken by mnemomap and webbrain that presents related concepts in a graph instead. The other part is to visually show the divisions of the search results into different categories so they can easily be navigated through but also to give a quick overview of the subject, examples of that can be seen at e.g. mooter and kooltorch. Finally I would also like to mention kartOO that have, in my opinion, gone one step further and even presents the links to the search results with images and icons.

In conclusion one can say that the ability to graphically visualize the search result so that it is possible to get a quick overview of a particular subject can prove to be a very important feature in future search solutions. It would not only help users find what they want to know, but also help them get a better and wider understanding of a particular subject, without forcing them to read through a large chunk of (hopefully) relevant text.

The search result and related concepts can be presented graphically instead. That will also take advantage of the fact that people can take in a lot more information through an image then by reading text. Further it can help the user to easily see if he or she is on the right track and make possible refinements to the query even before any returned document has been read through, thus saving valuable time, which today is more important then ever.

Lifestreams and Google

Google recently announced their new experimental site and it holds a feature to see search results grouped on dates, visualized in a time line, based on extracted dates from the source documents.

Simple entity extraction isn’t that complicated, especially not when it comes to keeping track of dates – some regular expressions can easy detect date formats and normalize them for any search engine to keep track. However, other vendors have at the most visualized them as dynamic navigators or just used it as metadata . The key lies in visualizing it and providing a user friendly interface, which I must say Google, again, has succeeded with. This announcement made me think of something two researchers from Yale wrote an article about in 1996.

Freeman and Gelmter (1996) introduce a new metaphor, “Lifestreams” – an idea for users to organize their own personal workspace. The concept is that everything a person creates or are involved with are attached to a lifestream, which simply is a visualized time line linked to a storage repository. Later the user can add filtering and alerting functions to monitor and summarize their the streams for easier overview or just narrow their streams down to suit their current information need.

For example, if a user wishes to add a meeting, there is no need to open up the calendar, just attach the meeting entry to the life stream in the future. Adding the proper alerting, one will still be reminded. Lifestreams are not only historical, the concept also expands into the future. Furthermore, the concept grows to having various substreams, one for your private, one for your professional, dynamically as you create it.

Furthermore Google also announced their new map search, where one could search for a almost anything and get a rough overview (of course on a map) where places related to the query is displayed highlighted on the map. It’s noticeable that Google hasn’t utilized all their content in these two experimental features, but I must say I’m really looking forward when they do. I would say that Google now are really starting showing off that they definitely are capable to deal with true contextual search.

So, what about the lifestreams? Well, the recent buzz around Enterprise 2.0 and moving the Web 2.0 with social networking and blogging inside the corporate firewalls could really learn from these three concepts. If some vendor would provide functionality for ordinary businessmen to build personal lifestreams, hook their documents, meetings and resources to it combined with easy controlled vocabulary tagging and visually on a map attaching it to location. Let’s add the last magic spice: search to aid building up enterprise historical lifestreams and provide easy access, entity extraction, filtering and alerting one would have the silver bullet. One could follow the lifestream of an organization as one follow the heartbeat of a human being. Enterprise Lifestreams of aggregated deparment lifestreams of aggregated individual lifestreams. Imagine visualizing an organizational lifestream and tap into the pulse of an enterprise organization. Imagine to share professional lifestreams with collegues and interestgroups.

Not to take over John Lennon’s role about imagine, but I feel this could the next killer enterprise application that would glue all enterprise 2.0 concepts together. This could be the next step in social networking and truly support processoriented organizations’s everyday work! What do you think? Are our enterprises ready for something like this?