Models and Theories in Human-Computer Interaction/TAM vs. Diffusion at prediction success in enterprise CMS replacement

The TAM model of innovation adoption tends towards relying on analysis of the factors around an individual’s acceptance or rejection of an innovation rather than framing such analysis in terms of a group or groups’ reception of an innovation. In my experience that is not an effective or realistic approach, as the innovations I've witnessed being researched, predicted, developed and implemented do not affect an individual… they affect entire teams and divisions internally and subsume entire demographic groups within our external user base.

An example of this phenomenon in our organization can best be illustrated by looking at the elements of the Diffusion of Innovations model instead and relating them to the specifics of our technological and cultural environment. The innovation I’ll describe is a CMS (content management system).

We have relied for many years on an outdated and ill-fitted CMS platform despite a number of attempts (by individuals in power vs. by empowered and knowledgeable groups) to install replacement platforms. The communication channels used in the past were limited to a small group of executive members which resulted in isolation from communal knowledge, from input of subject matter experts and most importantly from a nearly complete lack of feedback and subsequent buy-in from stakeholder groups.

In each effort to acquire a new software platform or build one from scratch, the length of time needed to complete the innovation-decision process was impractically extended by the failure to visualize properly and design a solution for the relationships within the company’s social system - namely that of the sales, editorial, design and technology groups. This was a failure in two ways. First, the leadership underestimated the need to engage these groups such that their common goals were identified and mapped to formal software requirements. Second - the time needed to effectively gather requirements and implement a solution was at direct odds with the goals set for these groups in terms of revenue, service level agreements and standards of publication quality.

We’re currently in the latter stages of a recent effort at developing and implementing a modernized CMS which is showing great success and which has been successful in large part due to altering the baseline approach, i.e. the type of innovation-decision.

Past attempts seemingly relied on the Authority Innovation-Decision approach, relying on precedent-setting non-technology decisions made in the company’s past (primarily a newspaper legacy). These failed as already discussed.

The success we’ve seen in this iteration is due to the apparent adoption of the Collective Innovation-Decision approach, which although has been time-intensive and cost-intensive has proven to be effective at both ensuring the product itself is what is necessary to meet our business agenda and social group needs but also has allowed us to deploy the system to a surprisingly high rate of initial adoption and few instances of failure as defined by service outages, loss of revenue and employee frustration.

To summarize, the TAM model would not been nearly as useful as the Diffusion model in predicting the failure or success of the attempts we’ve made over the last decade to overhaul our enterprise publishing platform CMS.