Knowledge Management or What's a Website For?

I have done a lot of work in my career at the intersection of technology and social change. While I have never focused specifically on web development I have built many websites and most of them poorly. It has only been recently that I have been able to embrace the idea that less is more. So, I thought I would write out a quick post to describe what I see as the four types of websites that are differentiated by the goals that they serve. The first two are content sites and the second two are community sites. In reality, a real site may have aspects of all four.

Brochureware - Succinctly describe an idea, a goal and/or an institution

In general, the goal of a brochureware site is to provide enough information to generate a productive lead or catalyze the first step of a call to action. Attracting the right customers and filtering out the wrong ones is a great use of this type of site. A brochureware site clearly describes the value offered to filter out some leads and hone in on those you are best able to serve with your available resources. A brochureware site can be highly interactive with compelling ways to interact with content; however, a defining characteristic is that the site is quite static as very little new content will be added over time. You will see a lot of hype that brochureware sites “are dead” but mostly that comes from people trying to sell you some whiz bang. A Brochureware site can be extremely useful to provide a low touch front door for a high touch process or to present complex information in an intuitive way.

An example of this type of site is TaroWorks, a startup social enterprise at the Grameen Foundation. This website was very effective as a place to send our customers to gain more information as well as a front door for unsolicited leads. Another interesting example is Risky Business. This is a website that is designed as a destination for a report by the Rhodium Group on the Economic Risks of Climate Change in the United States. A similar site for the very excellent report on social mobility in the United States is the Equality of Opportunity Project.

Curated Content: provide ready access to relevant content

Academic research is an archetypical example of curated content and my favorite nerd playground - the MIT Media Lab - is good example. Their website has a simple list of each program and clicking through takes you to a page with the individual research projects and then links to each researcher’s site. The FSG Knowledge Exchange is another excellent example of the same idea. Other examples are SSIR,Next Billion and Markets for Good. While these sites have excellent activity via reader comments, the focus is very much on the curated content. These sites are essentially interactive periodicals. These sites are very good at disseminating information and their success metrics are classic web metrics of hits, downloads, comments, tweets, etc.

Given that the goal is for readers to find engaging content, the achilles heel of these sites is navigation and content discoverability. Explicit search and browsing via topic and date is the default. Once a site gets a lot of content, the process is more difficult. The best content recommendations come from humans who know both the readers and the content. This is best done by humans employed to do this work - community managers. It is also possible for readers to rate content or to leverage “user behavior” to create an Amazon type recommendation - “Readers who read/liked this item also read/liked”. Building a valuable site that makes content readily available is a significant accomplishment and in many cases is the best case scenario.

It is worth noting that there are many sites that are defined by a specific community but that are still squarely in the Brochureware and Curated Content categories. Some examples of this are Business Call to ActionBiomimicry Institute and Social Capital Markets. Each of these sites has a well defined community as their audience and the purpose of the site is to serve content to that community to advance the goals of the organization. These communities clearly interact extensively offline but the site itself is not facilitating the interaction of the community.

The next two categories are focused on community interaction - on how the sites users engage with existing information as well as generate new information.

Knowledge Management: capturing, distributing, and effectively using knowledge

Knowledge Management (KM) means different things to different people. For the purposes of this post I will use this definition: "Knowledge management is the process of capturing, distributing, and effectively using knowledge." In the Curated Content section above I only discussed ‘distributing’, knowledge management adds both ‘capturing’ and ‘using’ to ‘distributing’.

A Knowledge Management site is the first that I have discussed that necessitates a community - a group of people that are impacted by their connections to each other. A knowledge management site would provide the opportunity to not just disseminate information but to tap into the process of using and capturing information to understand what was done. To achieve this end, knowledge management needs to be embedded into process or project management.

This is a big (and repeated) insight for me - that management of a meaningful process is the goal and that activity by a community of people on a website is the means to that goal. With technology we constantly get stuck in the trap of technology as the goal. Activity on a community site is not a goal - the output of that activity is the goal. In this case, the activity is engaging in a process of capturing, digesting and implementing knowledge.

The simplest version of this is the question and answer process of a user community. The Foundation’s Power of Us community is a great example of a vibrant user community. I talked to the community managers at Foundation and they stressed facilitating a call and response rhythm where questions are asked and answers are provided by the community members. This community works because the context of the questions and answers is very clear - using the specific features in the specific community of nonprofit organizations. In this community, experts are available and incentivised via reputation enhancement (which may translate into new business) and experts are developed over time as it is an active learning environment. Because experts are developed ove time in the context of the community they are also motivated to "give back" bay answering questions they themselves may have asked before.

XDA-Developers: Technologists are uniquely good at this sort of activity and there is a long history of social change practitioners trying to borrow technologists methodologies (open source and agile development are two examples). What I am looking for are great examples of this actually happening. On XDA-Developers you have a fantastic give and take of information and "threads" that you can follow from request, to information provided to information used to report out of experience. It is definitely worthy of another blog post on this topic alone.

Also, if you ever hear yourself saying, we want to build the facebook for X or the LinkedIn for Y, stop building a website and use facebook or LinkedIn.

Peer Production and Learning

This final type is not really a website on its own but more a set of features that is designed to generate content or manage a process to get a specific body of work completed. The classic example is a wiki. Wikipedia is the obvious example and there are millions of other wiki’s around the web. Essentially, a wiki is a free-form editing platform that is good for very highly engaged communities that need to generate content collectively to achieve a very specific goal. While some wiki’s are persistent, it is more likely that a wiki is used periodically to address a specific need. The difficulty of a wiki is that they are never self-managing and needs constant “gardening”. However, if what you need is to work with a large group of people to create content on a complicated topic, a wiki is an excellent resource.

Another example of a targeted and periodic functionality is event management. Every conference has a website and in the last several years conference website have improved to provide the ability for attendees to declare what sessions you will attend and to set up meetings with other attendees and for speakers to upload content and interact with attendees before the session. Some examples of this are the SoCap Conference series that uses a product from PathableSoCap12 andShare. Personally, I have found this to be a good tool and to be very helpful to prepare for the conference especially as a speaker.

Finally, education and learning has seen dramatic advances recently and companies like Declara are using advances in machine learning to present the right content and the right expertise to the right people at the right time. (I tried multiple times to get a demo from Declara but one was cancelled and otherwise didn’t get a response.)

Recommendation #1 - Get content curation right first.

I think an excellent content curation sight is a huge win. Getting content curation right means you are paying close attention to discoverability and reader behavior relative to individual pieces of content and you can learn a lot about what is resonating with your audience. Commenting on an article is a really high bar and the content of a comment is pretty low value from the perspective of a database. #-of-Comments-on-article is a great value to sort by but the content of the comments can’t be discovered until AFTER the article is discovered.

I think a very important question is, how can you feed readers’ behavior back to your readers to help them make decisions? Was an article viewed/downloaded/saved? LinkedIn’s new blog feature provides this sort of feedback for views, likes and comments. You could put the # of downloads on each piece of content. The idea is to feed reader behavior back to other readers to help them find valuable content. Instead of comments (or in addition) maybe a net promoter score or “would you recommend” type question? What ever you can do to lower the bar for your readers to express an opinion relative to the content - especially an opinion that can be indexed and searched/discovered.

Recommendation #2 - Build the process and discipline to connect the offline to the online

This is a general comment and your business firewalls and privacy concerns will dictate how far you can go. However, the basic behavior is something like -employees should never send someone an article attached to an email when you can send them a link to a page with the article. If you believe that the process of logging in to your website is too much to ask then reconsider requiring logins for downloads. Your daily connections are the lead generation engine for your site. If you find that you can’t integrate the site in to some aspect of your daily interactions with your contacts then you should reconsider the design of the site.

It is worth looking in to how to connect your events more deeply with your website. Integrating functionality like Pathable can help people get the most out of a conference and, more difficult but possible, it can provide a bridge between conversations the started or deepened in the face to face setting and ongoing collaboration in the web community setting. This takes a very conscious effort to engage panelists or speakers after a conference but I think it is well worth trying especially if a specific request to connect more deeply for a specific reason was expressed in the face to face setting. I would not just generically ask all presenters to “be on” the site but instead, create a mechanism at the conference to express a specific desire to connect in an ongoing way. There is a very specific commitment implied and that is a good thing. The essence of the recommendation is to integrate the asynchronous somewhat less personal connection of the web community to the real time face to face interaction of a meeting. The caveat is that is must be done with a very specific project commitment in mind and a clear process defined at the conference.

Finally, webinars webinars is a good idea. Feeding back the data on attendance but also providing recordings. Putting on people’s profile that they attended the webinar. Webinars always generate great questions that the attendees happily transcribe for you! This is very valuable data. Feeding them back to the site is a good idea. Give the webinars a life after the webinar.

Recommendation #3 - Find opportunities for generative activity like peer production and learning

This is a new area in general and it is THE area I am most excited about. I have always been drawn to the promise of technology to help us work better together. My work at the Resilience Exchange is focused on this idea. Specifically, we are trying to create a behavior change for mission focused organizations to remix existing solutions before creating new solutions from scratch. Here are some more general examples of this sort of generative activity:

  • Major advances in online education have happened and there is a very real opportunity to provide formal online courses with real credits and certificates earned. This should be integrated in to your community sites if it is something you want to pursue.

  • Using a wiki or similar functionality to document an emergent process or methodology.

  • Explicitly create a best practice repository including a community derived taxonomy.

This would only be relevant if you had a specific and tangible output that you want to create to achieve a specific and well articulated goal. I think this is an opportunity for but it is a very non-trivial body of work.

The biggest takeaway for me from this analysis is that generic activity on a website is meaningless unless you are Facebook where all you care about is generic activity. What you want is to use your website as a tool to accomplish a goal.