Thursday, September 24, 2009

SaaS, Learning, Ant Colonies and Data:
Do we have an answer for businesses to improve the quality of their decisions?

Very few businesses exist today where the culture is entirely autocratic and/or patriarchal. Most businesses, engage in group decision-making (not necessarily democratic -- of course.)

This makes it very important for businesses to understand not only how individuals make decisions but also how groups make decisions. In turn, it implies that a business must:
(a) understand perceptions of risks of a group (especially to minimize group-thinking);
(b) measure performance of the activities and decision-making of the group/team;
(c) understand the individual's critical thinking skills and its impact to the collective intelligence.

Web 2.0, SaaS and great optimization techniques are available now for businesses. Finally! you can improve the quality of your team's decisions.

First goal of building a SaaS application: Make it affordable.
Without affordability, no one will buy or switch from traditional enterprise.

Second goal of building a SaaS application: Make it easy to use.
Without ease, workarounds will be created to reduce (Dare! I say) or sabotage the initial intent of deploying the technology. Also, an easy-to-use application typically reduces GIGO effects of data collection.

Third goal of building a SaaS application: Make it easy to both understand and learn from the collected data.
An intangible asset of a business is the immense quantities of data they gather. This data will remain a swampy mush to reckon with and not good clay to build with, if groups or individuals cannot easily extract and analyze the data. If they cannot easily (a) learn the risks (b) measure the quality of group-decisions (c) understand the individual's critical-thinking skills -- then they will continue to make bad decisions. This "ease of learning" also allows for collaborative sharing across the businesses.

Fourth goal of building a SaaS application: Easily create (almost automatically) the next iteration (version) of the application based on the learning and adaptation of the underlying business process it has modeled.

In my opinion, most products of most SaaS companies have not yet reached the maturity to meet this goal. If we can design and create applications, where the underlying data and learning changes both the underlying work process and the SaaS application then the value of SaaS will be much higher than most experts predict.

Recall, the slow-moving mudslide of traditional enterprise-wide applications such as traditional SAP, Oracle etc. What happens to your SaaS application, when users change their underlying work process for better -- how long will it take your developers to change and adapt the application?

Ant Colony Optimization techniques are becoming more mainstream. This technique uses heuristic (data-driven) models to help systems learn. The models operate on an extremely simple premise: Individual ant is not intelligent, but a colony of ants working together create a swarm or collective intelligence to do many tasks well. See reference [1] for more.

These techniques can be utilized to meet the third and the fourth goal.
Only when these goals are met, then the "cat coming out of the bag" event will happen for SaaS. See reference [2] for more.

[1] Lessons from ant colony overcoming biases web 2.0

[2] The Cat is Out of the Bag (Again): The Hidden (?) Business Model in SaaS

No comments: