locked
Business Input - High Adoption Rates vs. Universal Templates and "Acceptable" Adoption: False Dichotomy? RRS feed

  • General discussion

  • This is a SharePoint Development Issue, so apologies to those in app dev being taken on from scratch.

    After taking off some time to really dig into the issues of HOW you approach development and its relationship to how WELL the final product addresses the needs of individual business groups, I came away from the literature feeling vindicated:  after years of claiming that careful needs assessments, and in particular, working on site with departments was the only truly dependable way to insure you have system requirements that will be meaningful and welcomed by business groups taking ownership of the system.

    Having joined a new organization, I've run head on into a situation in which the head of the Application Architecture unit is clearly experienced and has a strong record of building applications for the company.  He was also the champion for bringing on SharePoint, but brought on a new System Architect and developer to relieve his workload.  The conundrum is his insistence that it is possible to build a single template that satifies the needs of the vastly different business groups that we are all aware make up any medium/large business.

    The claim is that this not only simplifies maintenance and deployment, but governance as well: a kind of 80/20 argument.   Posting articles and books on best practices that include business groups in the development process has not altered the discussion whatsoever:  I'm returning to the initial assumption to see if there isn't a way of looking at the applications, users, roles, and company organization in a way that actually does make it possible to create a single team site/collaboration site template that has enough variability built into the initial configuration to allow quick fulfillment for site requests, while meeting the goal of simplicity and common design elements that allows most divisions to recognize the core features, as well as specialized add ons.  My first reaction was "oh - not again." I'm certain that some core can be developed with end user selection of add-on parts and connections that might meet the criteria I've been handed. That said, this feels like it could easily be taking yet another stab at the same problem, with the most likely result being an expensive proof that the end result will be mediocre levels of satisfaction. I'm curious if others have tried to approach this issue with renewed conviction, finding specific areas where success is possible, and areas where success is highly unlikely.  "Collaboration" being tied to highly specific work processes, data types and required outcomes suggests that this is going to be an expensive proof that one size doesn't fit all.  Any thoughts or feedback will be most welcome.


    Bruce Byers


    • Edited by dansker34 Thursday, November 8, 2012 4:47 AM
    Thursday, November 8, 2012 4:38 AM