Information management

Doing research on a problem or it’s answers is just the start. Information needs to be sorted in a meaningful way to make it of any value for making better decisions. Especially for ones that affect more than just you.

Summary

When organising information, start with the audience in mind. Knowing what they are looking for can support the creation of conversations and documents which inform or inspire a decision for the better. In this post I will outline my current paradigm for storing and retrieving information, and identify which companies are doing this well to achieve good customer satisfaction.

Introduction

My grandmother bought me a new board game every Christmas until I became a teenager. One year it was “Guess Who” [1] which involves two people each picking a secret character from a set of options then asking if they had a certain possible attribute such as glasses, long vs short hair, shirt colour etc until you had deducted who the other person had chosen. This game encouraged me to deconstruct what I needed to prioritise from what may appear to be a random set of peoples attributes to get the the single piece of information I was after.

Learning about the dichotomy of an organisms features in biology [3], and the location of library books using the dewey decimal system [4] also assisted to get used to other peoples categorisation of things or information. It was Guess Who though which allowed me to take control of deciding the way information is filtered and on reflection was some of the best information management training I have received.

How to categorise information

Categorising data can identify trends and make it usable for a range of people. To store, filter and extract information from data I use a line of questioning similar to that used for open ended journalistic enquiry which tend to start with the following prompts: who/what/where/when/why/how. This is in line with my divergent approach to decision making and encourages a fuzzy approach to bringing together what could appear to be disparate pieces of information. It is important to have a clearly defined problem and what information you are seeking at the outset to make sure the information gathered is most relevant to the target audience.

Why take a divergent approach?

The wide range of individual beliefs and opinions means there is no universal way to categorise the physical, virtual and observed worlds in one consistent way. Every thing, idea, or experience has multiple aspects to it and the best way to consolidate them is to understand how it will be used. For example email is generally filtered by:

1) When it arrived

2) Who it came from

3) What is is about

Email management programs generally allow users to filter which of these elements is the primary way to sort incoming letters through a “sort by” option at the top of a column. After that users can create additional filters by creating folders within their inbox. These might relate to where a set of emails relate to (such as a town/department/event/organisation), why a set of emails were received (for approval/information/response/confirmation), or how it was formatted (eg from a shared folder/scanned document/transcript/notes/letter/attachment). There is also an option to colour code email which could relate to what the level of urgency is or another theme of correspondence.

The thing to note is that email is used for a variety of reasons and in some instances there may be a need to filter information outside of the first three levels which are default in email management programs. In my case I avoid using an email program as a to do list or archive as I dont think it is nearby robust enough as a way to store, retrieve and share information.

Private business finds out how to manage information about products or services all the time through market research. A robust filing or search feature makes it easier for customers to find what they want. Generally there are two or three elements which are the primary ways to filter data. 

An example of how to use a system for categorising things is eBay [5]. The problem eBay has solved is that people want to sell things they have to others who are looking for that exact item. To make the search easier eBay categorises their auction inventory by:

1) Where is the item used eg car, fashion, office, sports, etc

2) What type of item it is

3) How much it costs

The search field can be used to find what type of item you are looking for. After that results can be filtered by where it is located, what condition it is in, who is selling it, and in my case I am interested to hear why the seller wants to move it on within the product description. All of these are the main ways to categorise information about things for sale that a visitor to eBay needs in order to decide if they want to make a purchase or not.

In a similar way Apple [6] categorises virtual goods through iTunes where items for sale are grouped by:

1) What type of good it is (music/movie/tv/podcast/app)

2) What genre it falls under (for music could be rock/funk/pop/electric/dance etc)

3) Who created it (the artist)

Apple has reduced the need for people to compare prices by making only two or three tiers of prices. This could help to reduce consumers being overwhelmed by choice.

When categorising information that relates to observations of procedures, processes, principles, facts, concepts and structures the divergence in the way that people perceive the world comes to light. This is highlighted in organisations which are more service oriented.

Gartner [7] is a good example of this approach where they have a looser separation of the primary, secondary and tertiary level of information heirarchy on the front page of their website. The options include:

1) Who are you (“Insight for your role”)

2) Where are you from (“Insight to your industry”) 

3) Why are you here (“Insight on IT topics”)

What I like about this approach is that every piece of information on the Gartner website must have been tagged by each of the sepations within the three high level sorting fields to allow this feature to be a robust way to browse to relevant information. It reduces the likelihood that people will rely solely on the search feature and does some of the thinking for them.

A strong trend I have noticed as websites have evolved is that they have changed from being inward focused to instead be outward focused. In the early days many websites reflected the organisations hierarchy and left it to the visitor to figure out which part of the organisation they needed to engage with to get an answer to their question. Now though it is very customer focused (for the successful businesses anyway). I think this reflects a change in leadership from telling people what to do, to instead ask them how they can serve them best. If nothing else the online world is empowering consumers to expect the highest common denominator from industries around the world in relation to information management.

How I have put this into practice

From my perspective the more intentional filtering that is done with information, the easier it is to see trends, and make connections that may be less obvious. It is also essential to be able to pull a wide variety of information together and make it useful for identifying a problem, or how a potential solution may impact on that problem. When working with a team a robust way of putting information in and pulling it out can also save a lot of time.

The simplest way I put this into practice is designing electronic folders, naming files and including meta data to quickly find saved electronic information. I generally create about half a dozen top level folders according to where it relates to eg home, office, sport, social. Next I create subfolders associated with activities within each of those high level themes. After that I name files based on why the document exists, what it relates to and when it was created eg “Letter to John - invitation to meeting - July 2013.doc”. If I need addition filtering I can add meta data on an extra layer I might need to use to find something such as a persons name, or how I intend to use the information such as if it relates to a specific goal I have, or value.

It is interesting that the next Apple operating system “Maverick” [9] has created a clearer option to create meta tags for information which was considered one of the highest profile additions to the OS by PC Mag [10]. The tone of the author reflects my perception that search features is the way to overcome the cluttering that is happening in a lot of information aggregation services such as websites. It may also reflect the increasing popularity of hashtags to consolidate relevant information quickly on social media.

“What makes tags different from folders (as all Gmail and Evernote users know) is a file can have multiple tags, but it can only be in one folder. Some computer users are resistant to this change, as they've grown accustomed to the nested folder structure and in fact often build them to mirror their workflow. The good news for future Mavericks users is tags are an addition to the operating system, not a replacement for folders (yet). But mark my words, they are the future.”

The customisation of the IBM Watson supercomputer [11] to interpret written and even spoken word to direct people to relevant information is going to create a huge demand for coding of information for rapid analysis. I can see a need for government to do a lot of work before then to really listen to the public about what information they want from a specific department, and how they would like to be able to access it.

An example is my entry into this years HISA app challenge [2] where I designed an app which filtered Australian Commonwealth Government information contained within the Personally Controlled Electronic Health Record information from the patients perspective in order of:

1) Where the disease is located (superimposed on an image of the body)

2) What the disease is (description comes up when location is chosen)

3) Who provided treatment, when the consults occurred and what their advice was (details within the “team” list which is filtered by condition when the location is chosen)

Then to a lesser extend options were included for people to get information filtered by the above categories into:

4) Who can assist to follow the treating professionals advise

5) How to prevent or reduce the impact from a disease

6) Why a health professional advised a particular course of treatment.

Health care would have to be one of the most complex areas for categorising information. When designing a website for a public health organisation [8] the three main axis of data I was aiming for were:

1) When is treatment needed (emergency/recovery/rehabilitation/prevention)

2) What is the illness

3) Who is receiving treatment (age groups)

My intention was for additional layers of filtering to be available including:

4) Where the treatment can be received

5) Who is providing advice

6) Why particular advice has been given (links to governing bodies, non profits etc)

Faceted navigation was being explored as a way to bring all of this together before I moved on from the project.

Another way I have used this process is creating spreadsheets for actions lists and categorising feedback from stakeholders on a project I am managing. This allows me to use the “sort data” option to view feedback received according to who provided feedback, what it related to, where it was received, when it was received, and how to measure if it has been applied to the project.

I especially enjoy applying themes to feedback in real time with large groups. As a Community Engagement practitioner I set myself a challenge last year to facilitate a group discussion at a national science related conference and find out what the meaning of life is. It was a great way to guide the shared perceptions of dozens of people and agree on a way of structuring high level information in a meaningful way. You can read more about how I approached that event in another of my blog posts [12].

Where to next

If you work in a service industry and share a lot of information the best way to manage it is to ask who needs it, and why. From my experience I have found simply talking to people about how they look for information is a good starting point. This process warrants significant investment though as having good buy in from your team is needed to make an information management system work. For the ones that do there is a whole world of potential this can lead to for quality improvement project and ultimately giving the end user the best service possible.

Do you manage information for a large number of people? Feel free to share your strategies in the comments below.

References

[1] Guess Who game instructions

http://en.wikipedia.org/wiki/Guess_Who%3F

[2] “My eHealth” HISA App contest 2013 entry from Joel Penson

http://betterdecisionsblog.blogspot.com.au/2013/06/my-personal-app-design.html 

[3] Biological classification

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

[4] Dewey decimal system

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

[5] eBay

http://www.ebay.com.au

[6] Apple website

http://www.apple.com

[7] Gartner website

http://www.gartner.com/technology/home.jsp

[8] MLHD website

http://www.mlhd.health.nsw.gov.au/

[9] OSX Maverick website

http://www.apple.com/au/osx/preview/?cid=wwa-au-kwg-mac

[10] PC Mag - 10 things you should know about OSX Maverick

http://www.pcmag.com/article2/0,2817,2420287,00.asp

[11] IBM Watson supercomputer

http://www-03.ibm.com/innovation/us/watson/

[12] Finding the meaning of life, Joel Penson.

http://betterdecisionsblog.blogspot.com.au/2013/07/finding-meaning-of-life.html