Enterprise Architecture Tools & the Gartner Magic Quadrant
Gartner recently published their June 2008 Magic Quadrant for Enterprise Architecture (EA) tools. (See http://www.gartner.com/DisplayDocument?id=697707&ref=g_sitelink (registration required); or a free version at http://www.troux.com/company/news/pressrelease.asp?pr=080618_gartnermq_pr.xml).
A couple of things sprang to mind as I skimmed through the report:
Tools are rarely the main reason for unsuccessful EA activities. There's the commonly quoted phrase "a fool with a tool is still a fool" (and can add "a fool with a tool can do more damage more quickly").
The same is true for the Gartner Magic Quadrants. As a tool they can be misused, intentionally or not. The report has a specific scope, focussing on the modelling and repository aspects of the tools (and to some extent publication) as well as supplier capabilities. It's easy to be lulled into thinking that tool X is better than tool Y (even if it is only at a high level). But this ignores a number of other dimensions such as: the tool's ability to support different levels of organisational capability; how likely it is to be adopted by different user communities; how well it supports 'getting stuff used'...
Personally, I like the Gartner 'Magic Quadrants' . They're handy for showing people an overview of the main players (selected by Gartner, and less useful if there is something you want to promote that doesn't rank highly). The quadrants have a reassuring, semi-quantitative feel. And, they indicate 'ability to execute' vs. 'vision'. Surely not much more that people developing EA roadmaps need!?..
However, for many organisations mandating one tool is not the answer to successful EA. An EA tool portfolio and roadmap can allow different people and material to be brought into the 'fold' and provide a basis for integration. (This is where EA metamodels start to become handy, as well as adoption of standard modelling notations such as BPMN). There are several Enterprise Architectural maturity assessments around. Looking at the tooling needs for the different dimensions being assessed is one way of identifying tool requirements and priorities.
A couple of specific thoughts on the report:
It'll be interesting to see how IBM takes forward the Telelogic suite (and the access it now has to EA groups in various companies). And also how it deals with the potential synergies/overlaps/duplication between various modelling products such as System Architect, Rational Rose, Tau Modeller ... as well the requirements management tools (Requisite Pro and DOORS, though the latter has quite a mature following in the government and defence areas).
SAP: Aris (IDS Scheer) and Mega are still tempting for organisations with large SAP investments.
Microsoft isn't listed specifically (always interesting to see how Enterprise Architects react when you mention Visio). Visio is not a 'proper/traditional' enterprise architecture tool. However, many organisations do have models in Visio and most of the more sophisticated tools will support Visio import (and in some instances export) eg System Architect's improved Visio import capability makes it easier to leverage some of that existing collateral. Over the next few years, as Microsoft's "Oslo" takes shape; as the gap between design and implementation reduces, it will be interesting to see how things change. Portals (eg SharePoint and Visio 2007 +) will provide many organisations with a way of quickly providing interaction/discussion around models - outside of the traditional EA group.
Perhaps surprisingly (or not), there is no explicit reference to SOA in this EA tool assessment. Many of the tool vendors listed do provide SOA support linked into the broader EA capability to varying degrees.
Irrespective of what people think of the content of the Gartner EA tool Magic Quadrant many "decision makers" will look at this when considering tools and that is sufficient reason to at least skim the report.
Andrew
(As an example of a set of detailed EA tool features see http://www.enterprise-architecture.info/Images/EA%20Tools/Enterprise%20Architecture%20Tool%20Selection%20Guide%20v4.2.pdf )
Personally, I like the Gartner 'Magic Quadrants' . They're handy for showing people an overview of the main players (selected by Gartner, and less useful if there is something you want to promote that doesn't rank highly). The quadrants have a reassuring, semi-quantitative feel. And, they indicate 'ability to execute' vs. 'vision'. Surely not much more that people developing EA roadmaps need!?..
However, for many organisations mandating one tool is not the answer to successful EA. An EA tool portfolio and roadmap can allow different people and material to be brought into the 'fold' and provide a basis for integration. (This is where EA metamodels start to become handy, as well as adoption of standard modelling notations such as BPMN). There are several Enterprise Architectural maturity assessments around. Looking at the tooling needs for the different dimensions being assessed is one way of identifying tool requirements and priorities.
A couple of specific thoughts on the report:
It'll be interesting to see how IBM takes forward the Telelogic suite (and the access it now has to EA groups in various companies). And also how it deals with the potential synergies/overlaps/duplication between various modelling products such as System Architect, Rational Rose, Tau Modeller ... as well the requirements management tools (Requisite Pro and DOORS, though the latter has quite a mature following in the government and defence areas).
SAP: Aris (IDS Scheer) and Mega are still tempting for organisations with large SAP investments.
Microsoft isn't listed specifically (always interesting to see how Enterprise Architects react when you mention Visio). Visio is not a 'proper/traditional' enterprise architecture tool. However, many organisations do have models in Visio and most of the more sophisticated tools will support Visio import (and in some instances export) eg System Architect's improved Visio import capability makes it easier to leverage some of that existing collateral. Over the next few years, as Microsoft's "Oslo" takes shape; as the gap between design and implementation reduces, it will be interesting to see how things change. Portals (eg SharePoint and Visio 2007 +) will provide many organisations with a way of quickly providing interaction/discussion around models - outside of the traditional EA group.
Perhaps surprisingly (or not), there is no explicit reference to SOA in this EA tool assessment. Many of the tool vendors listed do provide SOA support linked into the broader EA capability to varying degrees.
Irrespective of what people think of the content of the Gartner EA tool Magic Quadrant many "decision makers" will look at this when considering tools and that is sufficient reason to at least skim the report.
Andrew
(As an example of a set of detailed EA tool features see http://www.enterprise-architecture.info/Images/EA%20Tools/Enterprise%20Architecture%20Tool%20Selection%20Guide%20v4.2.pdf )
Comments
Hi Andrew,
Like your analysis of the recent Gartner report on EA tools. I saw it just a few days go, though not so analytically.
Agree on your comments on suitability of an EA tool for various organisational capability levels. In one customer exercise, we selected Casewise an EA tool because of its suitability for occassional user.
While use of metamodels nearly eliminates need for standardising EA tool within an organisation, there are other pragmatic reasons for minimising tools.
I have found PlanningIT from alfabet quite fascinating, though have not used it yet. It has moved from being purely visionary in the last report to a new entrant in this report's magic quadrant. Its strongest point is implementation of workflows for the common use cases of EA - viz. strategic planning, investment reviews, architecture reviews, technology infrastructure reviews and so on. It nearly hides the EA. I will like to know if anyone has used it on an engagement.
Posted by: Rajeev Arora | July 24, 2008 3:44 AM
Andrew
A very interesting viewpoint indeed. I am working with a client right now that has a similar dilemma. They are leveraging an Enterprise Architecture tool for communicating EA to the wider community. The challenge here is that not all EAs in the group have agreed on a common nomenclature and means of depicting their views, leading them to point fingers at the tool. I guess it is a part of my job to bring the common views and take the focus off just the tool.
Cheers
Posted by: Mohan Babu K | November 20, 2008 3:45 PM