To understand what inverted search is, it’s helpful to define what search is. In search’s simplest form the user has an idea of something that they are looking for. They type in a static query on a site hoping they get lucky. They are likely trying to pull some info that is just past their finger tips.
Inverted search is about designing a system that pushes out info that a person may not have been aware that they needed to know about. It’s a process of sending things dynamically at the right time, through the right delivery systems and on topic.
Other factors such as time and past events creates context to makes the info appropriate to take action with. The challenge with inverted search is that there needs to be some sort of setup ahead of time to define a range of interest. It’s tough to create a system without any starting info. Equally difficult is asking someone what their unknown interests are.
There’s a couple approaches for the system to extract the appropriate information to define boundaries. One is to set up a range for a number of topics. For instance if someone wants information on weather and how that is affecting travel, chances are they might not be interested in sports for the same issue. If the person can turn off that topic they won’t be sent info they have no interest in.
There’s another option though. What if the user can’t really define what their interests are going to be? In fact they don’t want to miss anything and want to consume everything. It’s an exploratory phase that allows the user to gauge the amount of information coming in, and what they can do with it. As they receive the info based on the delivery system they’ve chosen they can understand what range of topics they’re interested in.
The challenge is to be as informative as possible with the info they are getting in terms of why they are getting this. The second challenge is to make it easy for the user to modify a particular topic by either increasing the amount or decreasing the level.
The system also has to define a threshold to pass along important info on a topic even if the user has decided not to get information.
Trying to define a user cycle in terms of novice, intermediate and expert is difficult within the context of inverted search. Someone starting off may be overwhelmed with too much info while an expert may have taken the time to fine tune the system to fit their needs in a way that the novice hadn’t considered.
On the flip side the user just starting out may select a narrow set of options that doesn’t deliver many (or any) results. The person that has spent a lot of time within the system and understands the nuances has probably tweaked their thresholds. The design challenge is to turn that novice into that expert in a short period of time.
Another consideration is how the user will want to see their info. Along with delivery is timing. Important info doesn’t necessarily happen during normal work hours. Along the same idea of timing is quantity of info. Depending on the importance people could want a certain level of importance sent to their phone while info that meets another threshold is delivered through email.
All these levels of delivery need to be considered in a way that the user can understand, modify and adjust when needed.