louisrosenfeld.com logotype

Home > Bloug Archive

May 03, 2005: Updated Enterprise IA Roadmap

EIA Roadmap thumbnail

I've finally found a little time to update my Enterprise IA Roadmap (57Kb PDF file). (The first version--46Kb PDF--dates to August, 2003). Although it's the basis for the design section of my EIA seminar, I'm hopeful that anyone who is dealing with the "silo" problem might find it useful, seminar or not. If nothing else, it's a decent straw man to get people thinking differently about how to organize information inside a large, distributed, and politicized enterprise setting. In other words, if the progression of steps I've laid out in the Roadmap doesn't work for you, at least you'll have a starting point to react to and improve upon.

Below are some basic about using the Roadmap. There is a lot crammed into those little boxes though; I'll be glad to elaborate in this Bloug entry, and hope to learn more from you so that the next version might be even better.

What's the Enterprise IA Roadmap? A diagram that helps information architects and other designers make their enterprise's content easier to find regardless of which department maintains it. The goal is to integrate content from across departmental "silos" in ways that make sense to users.

How does the EIA Roadmap Work? It breaks down the universe of IA design into four distinct tracks and numerous sub-tracks, and plots concrete steps within each track over time. Rather than suggesting specific phases, the Roadmap describes a continuum of actions, ranging from "Very Soon" tasks that can be performed by one or a few information architects with minimal resources to "Way Off" actions that require coordinated, funded EIA teams.

Use the EIA Roadmap to...

  • Break the large problem of enterprise findability into small, digestible, actionable steps that lead to realistic goals and quick wins
  • Identify and sequence specific ways to help users find information across departmental silos
  • Demonstrate that "silver bullets" (e.g., a search engine or metadata) are part of a bigger picture of interdependent steps that should be considered first
  • Make the case for the resources required to tackle broad EIA challenges, and utilize those resources more effectively

Don't Use the EIA Roadmap as...

  • A project plan; instead, use it to create a project plan for the near-term actions that your organization needs to undertake
  • A source of truth: instead, use it as a straw man that you should react to and modify to suit the needs of your own organization and its users

email this entry

Comment: Lyle Kantrovich (May 4, 2005)

So Lou, what changed, and why? Inquiring minds want to know. :)

Comment: Domingo Ruiz (May 4, 2005)

Care to explain this a little bit more? I'm lost on a lot of these boxes. It'd be nice if they linked to definitions or detailed explantions. Also, what do you mean by silos?

Comment: James Melzer (May 6, 2005)

What I would love to see is an example scenario using the roadmap. A coherent plan built from the techniques suggested in the roadmap for a real (or made-up) organization. Why pick one technique over another? If you pick one, what are a couple others that work well with it? And so forth.

Comment: jim wilde (May 8, 2005)

Hi,

Great content! You put a lot of work in to your EIA Roadmap. Thank you for sharing it with me. I like the idea of multiple paths ie, top, bottom, guerrilla for IA. Where do you see something like del.icio.us fitting in to your schemes?

I have a couple of pilots going on with a tool I developed - Ideascape. We are using internal and external blogs, bookmarking/tagging for km and a few other novel applications.

We also use both, a structured taxonomy and a loose folksonomy to find and discover information. Users love the tagging and note taking ententions in bookmarking but it is too early to tell how well this mix and match approach with work over a longer period of time.

Comment: Robert Wetzlmayr (May 10, 2005)

Concerning the guerilla information track...

The concept of k-logs has some flaws: The art of writing well is not equally ditributed among key staff members of large corporations. Neither is the urge to build an "ego brand".

How would one motivate the most knowledgable staff members to put *their most valueable asset*, long term acquired know how and deep insights, into the reach of the unwashed masses? Remember, todays economy is largely driven by the "survival of the fittest" dogma, not by altruistic values?

Comment: Lou (May 16, 2005)

Hi all, so sorry for not answering your questions; it's been crazy for me the past week or two. Some responses below:

Lyle Kantrovich asked: "So Lou, what changed, and why? Inquiring minds want to know."

Hi Lyle, here's what I changed:
* Combined the two bottom-up tracks (content modeling and metadata) into one
* Added a sub-track to Top-Down Navigation on main page design
* Added a sub-track to Bottom-Up Navigation on metadata tagging
* Split the Search sub-track on interface and query design into two sub-tracks
* In just about each sub-track, I made some modifications, some major, some minor

Domingo Ruiz asked: "Care to explain this a little bit more? I'm lost on a lot of these boxes. It'd be nice if they linked to definitions or detailed explanations. Also, what do you mean by silos?"

Domingo, I'd considered defining each concept, but ran out of time and space. I use this diagram as a teaching aid for my EIA seminar, and tend to explain many of the concepts in the context of the seminar. Many of these concepts are fairly well-understood in the IA world at this point (and covered in IA books like the polar bear), but if there are specific ones I can help with, please let me know.

Oh, and you did ask about one in particular: silos. Silos are content areas (sites, sub-sites, portals) that grow organically around an enterprise's business divisions. In other words, silo-ized content generally corresponds to a business unit or department.

James Melzer asked: "What I would love to see is an example scenario using the roadmap. A coherent plan built from the techniques suggested in the roadmap for a real (or made-up) organization. Why pick one technique over another? If you pick one, what are a couple others that work well with it? And so forth."

Great idea, but probably outside what I can accomplish time-wise. As far as combining tracks, I think they're all present to some degree or another, but some more at times and some less. That's why the bottom axis is so important. For example, in my roadmap metadata plays less of a role in the near-term, more later.

jim wilde asked: "Great content! You put a lot of work in to your EIA Roadmap. Thank you for sharing it with me. I like the idea of multiple paths ie, top, bottom, guerrilla for IA. Where do you see something like del.icio.us fitting in to your schemes?"

It really could fit well into Bottom-Up (by enabling navigation by users who'd already found themselves deep into a site's content; also by generating tags). As an emergent approach, it might fit to some degree in Guerrilla IA, where users are creating structures that, by definition, tie together content across silos (and I'd not really thought of social bookmarking this way in the enterprise context before, so I'm glad you asked!).

Robert Wetzlmayr asked: "Concerning the guerilla information track...

"The concept of k-logs has some flaws: The art of writing well is not equally distributed among key staff members of large corporations. Neither is the urge to build an "ego brand".

"How would one motivate the most knowledgeable staff members to put *their most valuable asset*, long term acquired know how and deep insights, into the reach of the unwashed masses? Remember, today's economy is largely driven by the "survival of the fittest" dogma, not by altruistic values?"

Heh. That's why I have HR incentives in the "way off" end of things; it is really hard to effect this type of change management organizationally. It's more likely that we'll have to wait for enterprise policies to evolve in this direction, rather than somehow dragging managers to the realization that knowledge sharing is something worthwhile and, in fact, to be incented. Some enterprises are starting to figure this out, but this development may not be more than a few months old at this writing.

Great questions and comments; thanks!

Comment: Sérgio Nunes (May 18, 2005)


I saw a "Delete Site Index" item. Could you explain a little more on this?

Thanks for all the work.

Comment: Lou (May 18, 2005)

Sure; the idea there is that site indices are transitional tools with questionable long-term benefit. improved search systems will generally be better at supporting users' known-item searching. And specialized indices (e.g., a product index) are easier to maintain and still quite useful when compared with indices that try to cover everything and the kitchen sink.

Comment: K8 Simpson (May 25, 2005)

Hi Lou

Love what you and James Melzer have done to further the concept of enterprise ia with these diagrams.

Question: Is your Content Modelling strand about the connecting and mapping of shared terms and values (etc) between the various systems of an organisation? Something along the lines of what Andy Schriever talks about here: http://www.boxesandarrows.com/archives/the_knowledge-model_driven_enterprise.php perhaps?

If so this is exactly what I am currently trying to get my head around! Do you know of any resources out there (non-biased if poss.) that discuss the advantages and disadvantages of an enterprise system (all-singing/all-dancing) vs. a simple(-ish) relational database?

Comment: Lou (Jul 18, 2005)

Hi K8, and sorry for the delayed response. Yes, I think we mean the same basic object-oriented concepts with the term "content modeling". I see content models simply consisting of content objects, metadata (attributes and values), and links between objects (which may be driven by metadata or by other sources of connection, such as collaborative filtering data).

I don't know of any specific resources that compare enterprise systems with standard RDBMs. I'm not sure it's a fair comparison, actually; one model is far more distributed, and the content in an RDBMs is typically data, and doesn't include semi-structured text.

Hope this helps!

Add a Comment:

Name

Email

URL (optional, but must include http://)

Required: Name, email, and comment.
Want to mention a linked URL? Include http:// before the address.
Want to include bold or italics? Sorry; just use *asterisks* instead.

DAYENU ); } else { // so comments are closed on this entry... print(<<< I_SAID_DAYENU
Comments are now closed for this entry.

Comment spam has forced me to close comment functionality for older entries. However, if you have something vital to add concerning this entry (or its associated comments), please email your sage insights to me (lou [at] louisrosenfeld dot com). I'll make sure your comments are added to the conversation. Sorry for the inconvenience.

I_SAID_DAYENU ); } ?>