Dirk Knemeyer  

Fixing an Old-School Site Controlled By a “Webmaster”

July 22nd, 2005 by Dirk Knemeyer :: see related comic

From research and scoping to implementation and management, the components of successful interface design span many specialties and practitioners. But often the most challenging problems are not about the design, they are about organizational management and people. Here is one scenario and potential solution, to assist in your implementation of successful interface design.

##The Lowdown:
__Organization type__: Large, brick-and-mortar, relatively early adopters who began moving onto the web in the mid-1990s

__State of their website today__: Poor. Reflects 1990’s-era website design, with unsophisticated graphics, little-or-no apparent formal information architecture, limited adherence to current standards or best practices

__Web workflow structure__: Centralized, typically page-by-page HTML-based publication

__Web ownership/control__: Variable, but often split between marketing/communications (content, look-and-feel) and IT (functionality/hosting)

##The Bottleneck:
The Webmaster. Often a mid-to-late career professional who led or was instrumental to the development of the organization’s original web presence, and continues to be the key internal locus of control over the front end today

##The Context:

For a long time, The Webmaster was seen as a hero in the organization. When the world was tentatively stepping onto this newfangled thing, the web, this person already had or learnt the key skills to get their organization up and rolling, at a comparable level of quality and speed as what was happening across the mainstream web. Taking the humble requests and wishful hopes of various internal stakeholders and turning them into real, live, accessible pages on the web - often with motion graphics and lots of colours and lots of cute touches - this person earned and proudly toted the title of “Webmaster.” People heaped them with praise. They were important, and the key to unlocking a world that people were slightly in awe of and did not totally understand.

Thanks to the early successes of The Webmaster, more and more people and areas of the organization began stepping forward with requests: new sections, more photos, more graphics, more frequent updates, content - content - content! The Webmaster handled as much as they could. The organization did not properly add staff or resources, or manage workflow in a way that made the job manageable. Eventually, The Webmaster had to prioritize. Requests took a long time to fill, or never got taken care of at all. People started to get irritated. The Webmaster started to play favourites. Organizational web publishing became functionally broken.

While all that was happening, the Webmaster did not sufficiently upgrade their knowledge and skills, or evolve the content and quality of the interface at a pace commensurate with the rest of the web. The site looks completely outdated, and the usability is significantly worst than much of the rest of the web. Not only is the internal publishing process broken but there are deep divisions between the business units and the web publishing hub and the site itself is in an extremely poor state.

##The Solution?
There are a lot of issues that need to be dealt with to properly turn this situation around:

- __Personnel__: Fairly or not, the Webmaster is often considered the nexus of the problem, seemingly justified by the state of the website, poor service to various stakeholders, and a defiant attitude about the situation. Moreover, they have a strong attachment to and investment in their place and stature in the organization. A successful solution necessarily will decentralize web publishing, in whole or in part, through either bolstering the centralized resources and diffusing control between various web professionals, or transferring responsibility and tools to other areas of the organization, and changing the position of Webmaster to one of oversight and support, as opposed to total control. This change can be difficult for that person to participate in. They need to be included and given every chance to succeed, but it requires a degree of ego sublimation and healing relationships with various people and areas that can prove difficult. Management will need to be supportive of the Webmaster and the process, giving them every chance to succeed. Remember: many of the problems they are identified with today are as much organizational issues as they are problems with that employee. Show empathy. Work together.
- __Workflow__: As mentioned above, the workflow needs to be redistributed. Depending on the resources available to the organization, there either needs to be upgraded and enhanced staff added to the central unit, or a shifting of publishing responsibilities away from the centralized unit to different parts of the organization – along with the tools and training for those areas to be successful. It also requires an organizational management structure that is ultimately overseen by someone with people management skills and abilities, not someone with specific technical and communication skills, like the Webmaster.
- __Redesign__: The actual design of websites is the part most of us are most familiar with. If an organization is solving their workflow problems through adding talented staff, they may be able to handle the redesign process internally. If they are not adding key resources, or those resources are not independently sufficient to handle a major organizational redesign, you should work with an outside company to implement a thoughtful, successful, best practice redesign.
- __Application Technology__: Often, these institutions have an inadequate IT infrastructure and are just as wanting for a technology upgrade as the more visible parts of an upgraded web experience. However, provided that there are not upgrades necessary to enable key networked business tools such as an intranet or enterprise applications, this should be treated as a second order of priority. The most important objective is servicing customers and stakeholders, and this typically begins with the interface.
- __Tools__: Depending on the size and content needs of the organization, it may make sense to implement some type of content management solution (CMS). This will greatly simplify the process of making updates to the website, and help to ensure consistency in look-and-feel despite some degree of decentralization to the publishing process. It also puts organizational information in a more accessible form than as part of HTML web pages, for possible repurposing and use as appropriate.
- __Training__: Any type of decentralized publishing and workflow redistribution must include thoughtful, dedicated training to easily instruct people on the use of the system and participation in the process. This will take time to create and ongoing management to deploy and reintroduce as staff turns over or shifts their responsibilities.

##The Result?
Even following a thoughtful process, there are no guarantees. In my experience, success or failure is something measured along a continuum of greys, not a simple black-and-white. Most companies muddle through their interface problems and feature web interfaces and applications that are better than they could be, but not as good as they should be. The goal of a best practice web infrastructure that maximizes revenues while minimizing costs remains the ideal, but achieving that is almost an inherent impossibility. Particularly when people issues are a complicating factor, realizing a successful development process is a challenge. But even when we are faced with daunting organizational and people challenges, like the particular example explored here, following a rigorous, objective plan that considers all of the many facets that go into organizational web experiences will provide the best opportunity to do good work and make a meaningful difference.

_Dirk Knemeyer is a founding principal of Involution Studios LLC, based in Silicon Valley and Boston. He is a member of the Board of Directors for the , the AIGA Center for Brand Experience, and the Executive Committee for the User Experience Network. He publishes articles and audio programs at his personal website, www.knemeyer.com._

5 Responses to “Fixing an Old-School Site Controlled By a “Webmaster””
Michael Clark wrote:

Excellent article. Definitely sounds like significant portions of my day job.

Easy Question: You have the title webmaster in quotes. What job title should be used now for the person who manages the CMS? Web Manager? Web Editor? Online Publications Manager? etc….

Hard Question: At what point should the webmaster move on? Yes, I have a high degree of ownership of the web sites I manage. When is enough enough?

Thanks, Mike

james wrote:

At the institution I work fo I am referred to as the “Web Administrator” and my role mixes to a certain degree with our other Netadmins since I deal with web server health issues. My primary task has become, since we implemented ContentXML as our CMS, to administer it, handle programming issues related to our web site and coordinate design elements. My job is still somewhat of a blending of multiple specialties but that is the MO of most religious institutions, especially schools. We did however used to be very much like the bad scenario described in the article except with a higher level of decentralization since we basically used many contributors with FrontPage who managed individual site sections…OH how glad I was when shortly after I came on board they began to search for a better way that didn’t include FrontPage.

dk wrote:

Thanks for the kind word, Mike.

RE: job title, I think it needs to be organization specific. That is, I don’t think there is one title that is universally appropriate for the person who manages the CMS because, depending on the size of the organization, the size of the staff related to the website, and the political/titling structure. For example, the use of the word “Manager” as opposed to “Editor” has pretty significant implications in most organizations. The trick is representing the tasks the person does as well as their relative authority and scope in the organizational structure.

RE: moving on, while it depends on the individual, in my experience, both the webmaster and organization would benefit from a fresh start. There are typically too many fractured relationships and lack of faith in the webmaster/process for that person to ever be fully recognized or empowered for the skills and contributions they do have. And, as is typically the case any time someone goes from a position of great power to one of relative normalcy in the same organization, it is a tremendous struggle for that person to reconcile this change from an ego and pride perspective. On the other hand, the very act of that person moving on and someone new and enthusiastic being infused can have a resonant effect on everyone related to web publishing. And from the webmaster’s perspective, by finding a new job with a new company - even if in a superficially “lower” position - the enthusiasm over their hire and contributions can do wonders for their outlook and self-esteem. This is all based on the relatively dire scenario outlined above, that I’ve encountered a few different times. There can certainly be exceptions, of course, and your personal situation could be far less toxic than these tend to be. I would be happy to give you more personally directed feedback on your situation if you email me privately.

dsandler.org wrote:

Httpd habilis (or is it Apachepithecus?)

Today’s OK/Cancel comic features a cameo by not only David Siegel’s seminal (and now somewhat outdated) Designing Killer Websites, but also the “soothing green light” Slashdot t-shirt. Thanks, Tom!…

Xopl wrote:

There are two problems with this.

A) We’ve actually been TRYING to decentralise content publication, but it turns out departments cannot be trusted to keep their own content up to date.

B) Consistency and quality begin to fail, and it becomes somebody’s job to fix things all day long.

There needs to be a central team of designers, programmers, system operaters, a communications person/editor, and then a team manager with vision to hold everything together. Content publication can be decentralised if the input is simplified and semantic, with design concerns separate. The designs can control quality where the previous fails. The communications person can press your publishers to keep publishing.

Leave a Reply

OK/Cancel is a comic strip collaboration co-written and co-illustrated by Kevin Cheng and Tom Chi. Our subject matter focuses on interfaces, good and bad and the people behind the industry of building interfaces - usability specialists, interaction designers, human-computer interaction (HCI) experts, industrial designers, etc. (Who Links Here) ?