Design Strategy

To design a system to answer our three user needs, we developed 6 high-level design strategies.

These strategies involved conducting observations from our summative and formative research, integrating product and market needs as explicitly stated by Sun, and principles of good design.

Work within email

Our investigations revealed that email is the most popular location for task management and reminder systems.

Personal Reminder Systems

Source: Aggregated CI and P0 Surveys (n = 26)

The next most popular are sticky notes, which are often placed in persistent view. Other software includes Excel spreadsheets, text files, and electronic stickies.

Since our goal is to help people remember what they have to do, we recognized that we had a fundamental decision to make: to work within the existing structue we observed, or to attempt to modify it, and therefore introduces a new element of potential complexity.

We decided to embed real support features for reminders right into an email client.

Hypothesis
People will be more likely to use features located where they already go. This will lower the threshold required to both find new features, and the cost to change applications.

Explicitly declare tasks

The single greatest issue in the workplace we observed involves coordination failure due to lack of clarity surrounding who will do what.

We found this surprising, considering that we also observed 55 of 56 users* employed some kind of personal task management system. So what was the source of the problem?

Looking more closely, we observed that while personal reminder and task management systems are common, these systems do not share information. Much of the information stored in a personal reminder system is invisible to the groups an individual is working on.

We determined that one way to solve this problem would be to explicitly declare certain items as tasks, and to make those items both traceable and visible to all members of a group.

Hypothesis:
Public task statements can clarify ambiguity without significantly or negatively affecting working relationships.

* Aggregated statistics from Contextual Inquiry, and prototype testing

Lightweight structure

We observed that both corporations and individuals often commit to the high costs of planning when tasks are sufficiently high-risk. There already exist many different software systems to assist in such large scale planning activities, but our observations also identifed an underserved location for task management. There is very little coverage of the informal workflow. Interestingly, this is also a place where we observed much failure due to lack of coordination.

Rather than serve the managers and planners, we determined that a task management system might be applied to low-level tasks like group editing or signoff.

So we determined to fit lightweight structure onto this level of informal workflow. We would need to find a very low-cost way to apply just enough task-related infomation for the purposes of computation.

Hypothesis
Informal workflow will support a lightweight structure.

“What's in it for me? ”

CSCW systems need to account for how to motivate users.

Computer-Supported Cooperative Systems (CSCW) fail due to the “disparity between those who will benefit from an application and those who must do additional work to support it.”
Jonathan Grudin, CSCW 1998

People should get satisfaction by marking items as completed and there should be incentives for both senders and receivers of tasks.

Hypothesis:
Creating incentives for senders and receives can provide sufficient benefits to offset any additional work required of each group.

Incrementally innovate

Incremental changes to behavior also follows Sun's charge to deliver an advanced-development, rather than a research product.

Hypothesis:
Changing human behavior as little as possible will enable users to focus on the key benefits they will get from the few areas where we will ask them to change their behavior.

Centrally locate information

If we could locate all the information related to a task in a single location, we will save users search and retrieval time. This would mean creating a shared space within the email client itself that would house all the documents and information related to a task.

Hypothesis:
A single place to find information is better.

Product Demo