What is Information Architecture?

I want to continue to build on the theme of Information Architecture which is being talked about a great deal at the Open Group Conference in Newport Beach. In my post, “A Quick Look At The Importance Of Information Architecture” I highlight the value of Information Architecture and put it into the  context of Enterprise Architecture. 

In this post I want to define it and continue to build on that context setting. The area of information architecture is still a bit fuzzy on what it really is. I think the confusion starts with the name. Is this topic called “Information Architecture” or “Data Architecture”? Once you decide on a term you like, typically off to Google you go for a definition on wikipedia or some other site(s) that will contain variety of different insights into the terms.

I have my own definition of what Information Architecture is, and yes, I locked in on what I prefer to call this aspect of Enterprise Architecture. But let’s take a step back and look on the web at some definitions to see if there are some definitions that resonate.


Data Architecture

  • TOGAF Data Architecture – A description of the structure and interaction of the enterprise’s major types and sources of data, logical data assets, physical data assets and data management resources
  • Wikipedia – Data architecture in Information Technology is composed of models, policies, rules or standards that govern which data is collected, and how it is stored, arranged, integrated, and put to use in data systems and in organizations. A Data Architecture is often the design of data for use in defining the target state and the subsequent planning needed to achieve the target state. It is usually one of several architecture domains that form the pillars of an enterprise architecture or solution architecture.

Information Architecture

  • EIM Institute – Information Architecture is the function of defining and using master blueprints for semantic and physical integration of enterprise data assets (e.g., enterprise data model, enterprise data flows). These master blueprints provide a clear definition of how the data is structured, collected, shared, maintained, and stored from both the IT and business community perspectives.
  • Wikipedia – Information architecture (IA) is the art and science of organizing and labelling data including: websites, intranets, online communities and software to support usability.[1] It is an emerging discipline and community of practice focused on bringing together principles of design and architecture to the digital landscape.[2][page needed] Typically it involves a model or concept of information which is used and applied to activities that require explicit details of complex information systems. These activities include library systems and database development.
  • Web Monkey – Information architecture is the science of figuring out what you want your site to do and then constructing a blueprint before you dive in and put the thing together. It’s more important than you might think, and John Shiple, aka Squishy, tells you why.


Looking at these and it is really two ends of the extreme. So which one is right? 

Neither in my opinion, but some are close and I don’t think these definitions do the space justice. I also want to be clear that information architecture is a lot like business architecture a few years ago. Not well defined, loose methods, models and a real lack of definition around roles. 

The definitions around data architecture seems loser to enterprise architecture, however, it is a very technical definition. One of which I am not a believer in personally. Just like with all aspects of architecture there is abstraction, I believe these definitions are correct but at a lower level of abstraction.

As you read the definitions of information architecture from most sites it is more centered around User eXperience (UX) rather than what EA’s think of it. The outlier is the Enterprise Information Management (EIM) Institute, which has a definition that is closer to what I think the definition is. 

I like the marriage of the EIM Institute + TOGAF (and some parts of the Wikipedia definition). 

So here is the definition I land on:

Information Architecture is an aspect of enterprise architecture that enables an information strategy or business solution through the definition of the company’s business information assets, their sources, structure, classification and associations that will prescribe the required application architecture and technical capabilities.


The core of this definition is switching from starting with the application architecture (and sometimes even the technology architecture) but rather to focus first on business architecture that will lead to information architecture and then the other aspects. Once you define the IA aspects your architecture will more reliably align to the business value realization goals. 

This is enabled by many different methods, models and tools. I will talk more about them in a separate post. 

0 thoughts on “What is Information Architecture?”

  1. I’d agree that that’s a valid description of IA from a conventional business-oriented view: the architecture of the information that the business needs to see.
    Compare that, though, to the descriptions of IA more typical in the web / semantic-web / user-experience communities: there’ it’s more about the architecture of how the information is structured, presented, gathered, how it flows from page to page and task to task, and made usable and useful for the various actors in an interaction scenario.
    As you say, there’s a middle-ground, or perhaps a meta-level that’s more about linking information with users and uses of information. My feel is that your final-definition is still too scope-specific: I’d suggest we need to look more closely at that meta-level before coming back down again to specific scopes.


  2. IMO, a definition of Information Architecture (IA) that doesn’t focus on decision making is worthless. The sole purpose of information is to enable a decision to be made. Period. It doesn’t matter if that decision is whether a piece of RNA should attach to a strand of DNA at a particular point or whether to offer a customer a discount on a purchase.
    Gregory Bateson famously said that information is a “difference that makes a difference.” The latter difference refers to a different decision. See the definition of the word “decide” (v): “influence or determine the outcome of.” Information is something that determines the outcome of some behavior or action. That’s why information and process are yin and yang: process is the sequence of actions, information is what determines the path taken through such actions.
    Accordingly, IA should be defined as something along these lines (adapted from the IEEE-1471 definition of architecture): The fundamental organization of decision making; embodied in its components, their relationships to each other, and to the environment, and the principles guiding its design and evolution.
    Note, for more on information as patterns driving transformation of other patterns see http://en.wikipedia.org/wiki/Information#As_an_influence_which_leads_to_a_transformation .


  3. Very good overview of what you can find over the Internet when searching for IA. Two years ago, when I started my new role as Enterprise Information Architect, I came to the same conclusion as you did. TOGAF was way too technical oriented -as always – and EIM much closer to my believes. When it comes to Wikipedia… What a joke!
    However, I did shape the whole Information Governance framework and organization of my company (as it was my mission) accordingly to your definition. It was what naturally made sense to me and I must say, the only you to make sure we embrace the whole enterprise actors.
    Return of experience after 1.5 years (starting point of the Information Governance implementation) is very positive. We have managed to bring business focus and avoid the technical to phagocyte the initiative. I won’t say it was an easy task, though, but we did manage and we even succeeded to break through the process oriented mindset that was deeply anchored in the company.


  4. Great conversation with some rich feedback and real world experiences! All of you have some very valid points that I think add to cracking the Information Architecture “nut”. Keep in mind that I am far from the authority in this area. I really do believe this needs more attention from the community and the standards setting organizations to do this justice. What I am sharing are my experiences from running an Information Architecture function through my EA team. It’s been a journey but I think we are at a point to share some of those learnings.
    @Enectoux Sounds like you have actually put this to practice. That is great. I have an Information Architecture practice under me in EA as well. I think it’s really important for you to do what you just did, share with all the EA community your real world experience as a practicing Information Architect. Everything from the role, to the practice you have created and the specific value that your company gets out of this function.
    I think it would be great if you could share more about your experience. If you want to guest blog here on my blog or if you want to write something independently and I can link to it I really think it would help the EA community. Let me know and would love to wire it up!
    @Nick Gall I like how you called out decision making. I think that any architecting activity needs to focus on the key decision that need to be made. In your words “Information Architecture (IA) that doesn’t focus on decision making is worthless”, I think this is spot on but I think it also applies to all things we do as architects not just information architecture. I think we need to be careful not to fall into the trap of overloading the term “information” like we do with other words like “function”, “capability” or even “solution”. In this context I am describing Information Architecture not information it’s self.
    I rather think of your points as part of the method for performing Information (and all other domains) Architecture rather than the definition of it. The reference to IEEE 1471 (deprecated standard see: ISO/IEC/IEEE 42010:2011) I think is a good one, but as that standard specifies, that is a description of architecture all-up not just one of it’s domains.
    Thanks for providing the “information as patterns driving transformation of other patterns” resource. It’s a good one. But with this I see this an architecture pattern that is leveraged within Information Architecture not Information Architecture it’s self. This is akin to what you see in TOGAF each phase (B – D) starts with “Select Reference Models … “. I think this would be a key resource for that step in TOGAF.
    @Tom Graves Thanks Tom! Great feedback.
    I agree with your thoughts on the UX areas. I tend to think that this is the next level of detail from architecture and more design. That’s at least how I’ve structured it with my group. It tends to work out well given that the outcomes, skills and deliverables are so different. But I think it would be a good exercise to explore.
    I would love to hear your ideas on this and the meta-level IA definition.


  5. Mike: I like your definition. It’s a bit of a run-on sentence, but it pulls together the key aspects that define IA: (a) part of EA, (b) derived from and driven by business, (c)constrains solutions. Not sure about whether IA enables solutions… because that bit seems to say that the reason we create IA artifacts is to drive or constrain specific solutions. But that is only part of why we create IA. Another reason is to get the business leaders to agree with ONE ANOTHER.
    I had the good fortune over the years of working with some fantastic IAs. I’ve had the opportunity to play the role of an IA only a few times. I hold Information Architecture in high regard.
    That said, some of the most effective results that I’ve seen from developing a consistent Information Architecture comes from the agreement between business leaders that happens when they realize that they have not been consistent! Simply the act of creating the architecture and then using it to teach others creates and drives a level of consistency that would not otherwise be possible.
    The structure itself then drives tactical decisions about how information is managed, secured, rationalized, and consumed. This happens LONG before decisions are made from that data.
    Nick Gall’s definition sounds pretty silly for someone who’s spent the last 30 years making information move through organizations like most Enterprise Architects have.
    At some level, information is used to make decisions, but most information supports other information which supports other information… only the top layer is used in decision making. The rest is critical to the delivery of products and services and the satisfaction of customers because it keeps mistakes from being made and drives effectiveness, not because EACH bit of data is used to drive decisioning.
    Good post. Thanks for sharing.


  6. I’d like to agree with a somewhat generalized version of Nick Gall’s point. I don’t think it’s just decision-making but any aspect of Organizational Intelligence, in other words sense-making, learning, communication, etc. Some people (following Vickers) may use the word Judgement.
    Many Information Architects seem to think that Information Management ends (fulfils its purpose) when some (correct) information is displayed on some device. They seem completely unconcerned about whether and how this information is used. But as I see it, if it’s not used/usable then it is not meaningful, and if it is not meaningful then it’s just data/noise.


  7. Hi Mike,
    I like the idea! I’ve just have to plan a bit to publish things in the right (consistent) order and find a bit of time as well, but it should not take too long for me, since the content is there, I’ve just have to put it in a way that it is “digest-able” by all.
    When it comes to the publishing platform (blog), thank you for the offer, but I have already my blog – http://enectoux.wordpress.com/ – so I will do it there. What we can do is to link our blogs, as I do with Tom Tetradian for instance ;). Done my part already, you are now on my “Enterprise Architects blogs” section.
    I will start to write the first article today… Let you know when it is ready.


  8. Nick M: “At some level, information is used to make decisions, but most information supports other information which supports other information… only the top layer is used in decision making.”
    Nick M, It’s decision-making (or if you prefer judgment-making) all the way down. 🙂 That said, I agree that a BETTER definition of Information Architecture (IA) than the one I cobbled together for a blog comment would include some mention of the prioritization of business decision making. The IA should focus on the most important business decisions and use that focus to scope which information needs the most attention from architects.
    In my 17 years advising hundreds of architects, I see broad, unfocused, generic definitions like most of those above leading to “boil the ocean” approaches to cataloging everything under the sun without any sense of what the “architecture” is supposed to accomplish. I’ve tried to hold back the tide of architecture-for-architecture-sake, but sometimes it feels like I’m the boy with his finger in the dike. 🙂
    Richard V, Judgment works fine for me!


  9. Wow. This turned into a popular topic. Fantastic to see such passion around this space.
    I would like to start off with some critical assumptions that I am making here. From reading some of the comments from Nick Gall and Richard I think we are overloading the terms a bit causing a disconnect. This definition focuses on the information architecture discipline not in defining information and not how information is used to make decisions.
    Critical assumptions
    #1 – There is a difference between the definition of specific information and the practice/methods of Information Architecture (IA)
    #2 – All architecture efforts (regardless of domain) requires as an input to have information to make good trade-offs and business driven decisions
    #3 – Judgement / decision making is an activity you perform to generate an outcome, in this case it’s a top-down, business driven architecture. I say architecture because good decision making needs to happen for for all architecture efforts (Business, Information, Application and Technology Architecture). This applies all the way down to technology architecture.
    @Nick Gall I haven’t heard anyone so far anyone suggesting that we do architecture for architecture sake. I think the professionals on this thread are way past that point. I can speak more for myself and a little for Nick Malik (because I have working experience with) here. Both Nick and I are practicing EA’s that have both advised hundreds of EA’s and are personally accountable and responsible for EA delivery and value creation efforts. We are sharing our insights from the trenches.
    I’ve personally have evangelized pragmatic methods for EA for years. Nick Malik has as well. He recently posted a piece on Agile EA.
    As far as the assertions on decisions. I don’t disagree (see the critical assumptions above) but I think you are over generalizing in your explanation. By defining a practice by an activity is so generic and applies to everything I think it is fundamentally wrong. Under this logic we would define every thing in life this way (ex: Stack Trading – the act of good decision making with the information at hand). We define these practices by the outcomes they produce.
    @Emeric Thanks. I will check it out and link back to your resources on IA. It’s great that you can share for the rest of the community. I will share what I can as well from my side.
    @Richardveryard Thanks for posting Richard. I agree with you on what you said:
    “Many Information Architects seem to think that Information Management ends (fulfils its purpose) when some (correct) information is displayed on some device. They seem completely unconcerned about whether and how this information is used. But as I see it, if it’s not used/usable then it is not meaningful, and if it is not meaningful then it’s just data/noise.”
    I couldn’t agree more! That’s why the definition starts with the ultimate goal in mind:
    “Information Architecture is an aspect of enterprise architecture that enables an information strategy or business solution…”
    The key here is enables an information strategy or business solution. Both of these have the end BUSINESS problem in mind. There is nothing in this definition that is technical in nature and there shouldn’t be. We are at the level of business information entities (BIE) not ERD’s representing the implementation of information into data. That comes much later.
    @Nick Malik Yea, it is bit of a run on isn’t it. I must been feeling a bit British with my grammar 🙂
    I agree with all your points and you pointing out that IA is more than solving specific solution needs but also applying to much broader areas as well. My last comment clarifies that in the definition that the information strategy bit is that focus you are looking for. If there is a better way of stating it I would love to hear how we can modify it.


  10. Hi Mike – thanks for the response. I, uh, got into something of a tangle around my somewhat-doubting response to Nick’s assertion that it’s _always_ about decision-making, so it’s probably best I back off from this one for a while.
    All I can add right now is that there’s something that I’m trying to get at that I sort-of know is there but can’t quite put my finger on as yet, let alone describe in a way that would make sense to anyone else. Yeah, whatever-it-is is still sort of at the meta-level, and yes, Nick’s point about decision-making forms part of it but very strongly does _not_ feel that it’s the sole core of it, and that sideways twist also comes into it about how the UX people use the IA term is sort-of-process and sort-of-architecture and sort-of sideways again. And something about story, too – that whole thing about architecture as the interaction between structure and story that I worked on during much of last year, as you know.
    In other words, still way too blurry to be able to come with anything definite yet… sorry! Let me let it brew for now, I guess? But thanks again for the useful/interesting-challenge, anyway!


  11. Mike: “There is a difference between the definition of specific information and the practice/methods of Information Architecture (IA).”
    Mike, I’m not following you. First, my definition was NOT about the definition of specific information. So I’m not sure why you’re making this particular distinction. Second, the quote sounds analogous to, “There is a difference between the design of a specific building and the practice/methods of Building Architecture (BldgArch).”
    True…but nevertheless, the practice/methods of BldgArch should be essentially focused on the design of specific buildings. Shouldn’t it? I am reminded of the famous dictum from one of the early written works on BldgArch: http://en.wikipedia.org/wiki/De_architectura : “Well building hath three conditions: firmness, commodity, and delight.”
    IMO, a good definition of IA should, like a good definition of BldgArch, highlight SOME ***specific*** goal or attribute of GOOD architecture. Better decision making (or as Richard V would have it, better judgment) is the explicit goal of my definition of IA. What’s the goal of yours?
    If De architectura had a definition along the lines of your definition it would sound something like this: “(Building) Architecture enables a building strategy or real estate solution through the definition of the company’s building assets, their sources, structure, classification and associations that will prescribe the required facilities architecture and technical capabilities.” Lots of words saying very little.
    I, and I think most people, prefer “The ideal building has three elements; it is sturdy, useful, and beautiful.” Likewise, I think most business people would prefer a definition along the lines of, “The purpose of IA is to design the business information needed for more reliable, effective, and rapid business judgment.”
    Thanks for helping me refine my IA definition in the course of this discussion!


  12. @Nick Gall: Again, I don’t dispute the importance of judgement or business driven decision making here. Keep in mind, in the first few words I talk about the outcome of IA being an information strategy and BUSINESS solutions, to do this right you MUST have good judgment and business decision making. I fear that this is starting to take the form of a philosophical conversation so I will be as concrete as possible.
    Very simply, I believe that this definition is way to general for a definition of Information Architecture:
    “The purpose of IA is to design the business information needed for more reliable, effective, and rapid business judgment.”
    There is nothing to really argue about the statement and I agree with the words but it leaves me wanting more. If I was an aspiring IA I wouldn’t know what to think about IA. I’ve asked several business unit leaders that are stakeholders of mine about your proposed definition and replied back with, “this is yet another typical motherhood and apple pie definition”. That immediately follows with, “OK, tell me what I get? I’m really good at business judgement. What specific value to I get?”
    The problem with this definition is it’s so abstract that it’s hard to argue with and at the same time leaves you confused and with more questions than you had in the beginning. It reminds me of the John Locke quote:
    “Vague and mysterious forms of speech, and abuse of language, have so long passed for mysteries of science; and hard or misapplied words with little or no meaning have, by prescription, such a right to be mistaken for deep learning…”
    – John Locke
    Again, I think the definition I have posted most certain needs tweaks but I do believe that it’s much closer to the mark of what business stakeholders, architects and all other stakeholder of IA are expecting.


  13. So a definition focused on ***decision-making*** is too abstract and too motherhood and apple pie to be useful to anyone? Perhaps you should tell that to Nick Malik, because, guess what?, that’s how Microsoft defines Business Intelligence:
    “Microsoft BI solutions … empower users to gain access to accurate, up-to-date information for better, more relevant ***decision making.***” http://msdn.microsoft.com/en-us/sql10r2byfbi-trainingcourse_sql10r2byfbi00_unit.aspx
    “One definition of BI that seems to make sense is the process of organizing and analyzing existing data in a way that helps you make better-informed business decisions.” http://technet.microsoft.com/en-us/magazine/gg981676.aspx
    So while one may legitimately criticize my definition of IA and being too BI-focused, I don’t think you can legitimately criticize it for being too abstract or vague.
    In fact Mike, you beat me to the punch! I was going to point out that YOUR definition was too abstract to be useful! It is so vague that the basic structure of your IA definition can be used for TA (Technical Architecture) and BA (Business Architecture) or any ?A. 🙂 For example, here’s your definition of IA and then the same structure spun for TA:
    “Information Architecture is an aspect of enterprise architecture that enables an information strategy or business solution through the definition of the company’s business information assets, their sources, structure, classification and associations that will prescribe the required application architecture and technical capabilities.”
    “Technical Architecture is an aspect of enterprise architecture that enables a technical strategy or business solution through the definition of the company’s business technical assets, their sources, structure, classification and associations that will prescribe the required application architecture and technical capabilities.”
    What have we said with your definition? Not much:
    * It’s part of EA.
    * Somehow the archtitecture will “enable” strategies and solutions. (How? No idea.)
    * Somehow the architecture will “prescribe” application architecture and technical capabilities. (Seems like a pretty IT focused definition of IA! I’d rather see my IA “prescribe” business processes and business decision-making, not JUST how they are mapped onto SW/HW.)
    Your definition is so abstract that I have no idea HOW IA enables strategies and solutions or prescribes application and technical architecture. ANY architecture (information, application, technology, even building architecture for goodness sake) makes the abstract claim that it enables strategies and solutions!
    Perhaps it might help to think about IA as follows: An IA, no matter how we define it, will be eventually realized as metadata. Why? Because the IA will be realized as information about an enterprise’s information assets. And information about information is metadata! If we think about the purpose of an IA as it relates to the purpose of metadata, perhaps we’ll make some progress in grounding the defintion in something practical.
    Consider Gartner’s definition of metadata: “Metadata is information that describes various facets of an information asset to improve its usability throughout its life cycle.” http://www.gartner.com/resId=1424022
    What I like about our definition is that it focuses on the PURPOSE of metadata. The traditional definition of metadata (‘data about data’) focuses only on what it is, not what value it provides. The Gartner definition of metadata zeros in on the WHY: to improve the usability of information.
    In fact, the OTHER IA (the one defined by web designers) provides a similar definition of IA: “Information architecture (IA) is the art and science of organizing and labeling data (including: websites, intranets, online communities and software) to support ***usability***.” http://www.iainstitute.org/documents/learn/What_is_IA.pdf
    So in summary, improving decision-making is a perfectly concrete goal for an IA, since it works perfectly well in defining BI. If one wants to add additional goals such as usability or findability (was the web design IA folk want to do), I’m amenable to that. But one way or another, the definition of IA should be short and focused on the PURPOSE of the exercise itself. Agreed?


  14. Here’s another description of IA that emphasizes decision-making (from IBM):
    “Enterprise information architecture: Systematically unlock the business value of information through strategies and models that deliver actionable, real-time information in context to enable better ***decision-making*** throughout the enterprise.”
    I like it even better than my definition–much punchier! It reinforces my main point that a definition of IA that doesn’t focus on decision-making is rubbish.
    I came across it because I got an interesting email from IBM’s InfoGov Community in my inbox a few minutes ago: “I Am an Information Strategist.” So now we go from Information Architects to Information Strategists (who apparently lead both Information Scientists and Information Architects). 🙂 Here’s a short excerpt describing the role:
    ‘I am an information strategist. My job title is new, but my challenge is ancient: I have to get the right information to the right people at the right time. If the information is in the wrong format or if it is hard to read, confusing, or conflicting, with sloppy interfaces, then I have failed. Endless lists and search results without relationships, connections, and meaning are not what my customers demand. They want clarity. And they expect interfaces to match the ease of use they are accustomed to on their tablets and smartphones. This is the twenty-first century, and my job is to deliver information that is enlightening and entertaining.
    My customers—both internal and external—need insight, so by definition I have to know more about their business than they do. That means I spend a lot of time sitting in business meetings, listening and asking questions to understand the structure and context of what customers are doing. I join sales reps on customer calls, participate in industry conferences, and use my extended network to understand emerging trends and ideas in very diverse forums.
    Information strategists combine data management, information governance, and data science techniques to deliver refined, fit-for-purpose, high-value information products and services across the entire information supply chain. This role is an organic evolution of many information-centric roles that have been established over the past two decades to engineer data into silos, govern its uses, and ensure application access. However, information needs today extend far beyond data engineering and industrial automation. Most enterprises are filled with people who are information connoisseurs in their private lives—and these people are frustrated that enterprise information systems produce industrial interfaces. They want at work what they get at home: intuitive interfaces and information that is updated often and is fun to use.’
    Sounds pretty exciting. For the full description see http://bit.ly/Xm8nkz .


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s