The 7 deadly CMS requirements

A 13 minute read written by Paul October 8, 2013

CMS logos: OpenText, Oracle, Drupal, WordPress, etc.

Our team works with a pile of different CMS platforms both proprietary and open source. We praise and curse all of them frequently and equally. However, for such imperfect products, customers sure seem to like buying them over and over.

Analysts like Gartner suggest that the Web Content Management industry is worth well over $1B, growing at roughly 10% per year (Gartner, 2010). Customers keep spending money on something, and it can’t just be a better rich text editor. What problems do people think the next great product is magically going to solve that keep them coming back to the CMS store?

Killer features, for real!

Deane Barker of Blend Interactive gave a talk on the changing role of the CMS at DrupalCon Portland earlier this year. He thinks that content management capabilities are now commodity features. Any CMS platform boasting about its “new” features is competing with a vast array of existing, often free technology available in the market. I agree.

Most CMS platforms do the same set of things. We change CMS platforms hoping for some meaningful improvement, but are soon disappointed by the very features that convinced us to buy the system in the first place:

  • “Easy to use” content authoring features lead to content bloat due to lack of governance
  • “Out of the box” plug-ins and features create a disjointed and dissatisfying user experience
  • “No customization needed” deployments get cheers from the IT department, but produce a web channel with no system design or strategy for growth.

The “killer features” of many CMS platforms don’t address the core problem: content management is a practice, not a technology. Technology automates good practice.

We bought a CMS! Again.

I’ve purchased “Sticky Fingers” by the Rolling Stones four times: vinyl (twice, I needed to find a copy where the zipper worked), CD, and iTunes. I seem to be perfectly happy to keep buying the same product over again in a slightly more convenient package. Maybe this is why I’ve ended up working with Enterprise CMS products. We keep buying the same thing over and over, hoping for a different result, when what we really want is a CMS platform that lets us get past technology and down to our core business.

Many organizations are starting to build mature content management practices. Now that they understand what being a web-first organization really means, they realize that their CMS platform or implementation can’t meet their needs.

The CMS industry has some big problems to solve if it wants to stay relevant. If you’re on the digital agency side, these are the seven problems your clients are going to (and already) have – you need to be ready to help them.

1. Give content authors a coach

The first job of a CMS is to help content authors make great content. But very few CMS platforms give content authors reliable, usable tools for measuring content quality – or even quantity.

Every CMS should be able to tell us:

  • How much content do we have?
  • What binary assets do we have? Where are they?
  • Is our content easy to understand and use?
  • What content are people using (and not using)?
  • Who are our website visitors?
  • How often is our content updated? Who is updating it?

Once we have these answers, we can measure the basic costs of our web operation. But, that doesn’t give us the information we need to improve the returns on our investment. We also need to know:

  • Who is creating our best content? Who needs more training?
  • What content isn’t generating a return on investment? i.e. what content is updated regularly, but isn’t being read or consumed?
  • How well does our language match the language our visitors use? Do our navigation terms, titles, headings, and page content match the words they’re using in external and internal search?
  • What language, layouts, and images get the best response from our audience?
  • How effective is our non-website content – social media, etc.? What risks are we incurring? What rewards are we getting?

The next generation of CMS products has to give us clear, accurate data on our content’s quality and effectiveness. Content quality measurement belongs in the CMS, so our data can help coach content authors within their existing workflow.

2. Give IT an enterprise-grade toolkit

Business wants everything on the web – network-only applications aren’t good enough anymore. There was a time that IT supported back office systems and Marketing supported the web. Now they need to be able to collaborate using a common toolkit.

In this new Bring Your Own Device workplace, your CMS platform has to work with many enterprise systems, such as:

  • Directory Services (ADS, LDAP) for authentication and single sign on
  • Customer Relationship Management for lead generation and communication
  • Enterprise Resource Planning
  • Human Resources systems for encouraging employee self-service
  • e-Commerce systems that meet the dreaded Payment Card Industry compliance standards
  • Portal frameworks for securely exposing back office systems through the web.

These are expensive systems that took years to put in place. People have created work processes and practices that they use every day. They are big rocks that can’t be moved without significant risk and cost to the organization. A CMS is an inexpensive, malleable project in comparison. So if something’s gotta give (and it usually does) it’s going to be the CMS.

Most CMS tools aren’t flexible enough to work around these other enterprise systems and have a content author environment that people can actually use. Most CMS tools aren’t powerful enough to present these enterprise systems securely while supporting good user experience patterns.

3. Clear the compliance hurdles with ease

Big customers can be disorganized, but not on purpose. It’s hard to keep everyone on the same page. Here’s why content creation can be so complex:

  1. Everyone has to write in a consistent voice, use a common dictionary and style guide, and follow best practices for web writing and SEO.
  2. We’re creating content for the responsive web. Responsive data tables, video embedding, and mobile accessibility aren’t straightforward.
  3. Now add multiple languages to the mix. Many content creators don’t speak the same language or understand the same cultural idioms.
  4. Brand assets have to be strictly controlled and properly localized for each market.
  5. There are often federal or legal regulations to deal with - i.e. companies that advertise financial products.
  6. Oh, and content authors are usually in different locations and time zones.

And once we make it through that whole process, somebody has to accept all the risk of “approving” this new content. We have to help that person make informed approvals.

Most larger CMS platforms have robust workflow capabilities. But many of them don’t help decision makers to truly understand the context of their decision. For a financial services company, some products cannot appear without specific disclaimers in the footer of a page. If we are reusing content throughout the site, then the approver must not just approve content changes, they must approve them in each context that they appear, to ensure that the end page contains all the appropriate legal building blocks - even in a mobile view.

CMS vendors need to help content creators understand how their content will be viewed by the end user, so they understand the context of their work. This is in direct opposition to the next issue, which is that it’s difficult – and many would say undesirable – to demonstrate context to the author.

4. Admit that WYSINWIG (What You See Is Not What I Get)

The attempt to create WYSIWYG (What You See Is What You Get) environments for web publishing systems traps content in a specific layout and context. What You See Is Not What I Get should be the new mantra of CMS creators because as soon as content is entered according to a specific layout - i.e. desktop - it becomes partially irrelevant for other layouts such as mobile.

A simple example is a data table that displays a list of first and last names, mailing addresses, and phone numbers. On a desktop display, this may present well, but on a mobile device, the table rows may mean that a user can’t see the name of a person and their phone number together in the same screen once they zoom to a sufficient level of readability.

Daniel Jacobson’s great article on COPE explains that content needs to be modular and portable. It’s his opinion that we should create a new software concept called a Web Publishing Tool (WPT) whose role is to determine how content is presented on the web. According to Jacobson, a CMS should manage content and only content, not the “wrapper” of a website.

I suspect that there is conflict between what’s best for content and what the typical senior manager signing off on technology and service purchases considers to be best for the organization. A senior manager responsible for marketing is going to be focused on immediate business outcomes, like “we were able to create a web property quickly” not “over time we were able to abstract all of our content into a well-structured, presentation-agnostic state”. It’s going to take a new generation of content champions and leaders with a vision for the future of the web for more organizations to adopt the concept of the CMS as a tool to manage content, but not to manage its presentation.

The immediate problem is getting actual human being to create compelling, effective web marketing properties. The long view problem is creating permanent content repositories that abstract content from the specific web property so that we can stop building and rebuilding new websites with the same old content.

The tension between abstraction and contextualization is going to be the driver for most of the innovation in the CMS industry in the next five years. Companies like Adobe are betting heavily on this through their acquisition of both Day (abstraction) and Omniture (contextualization).

5. Make the CMS work at cloud scale

“Cloud” is the marketing pitch of the year. Google is making headway with email, office, and enterprise search delivered via a non-premised model. Microsoft Office 365 provides “anywhere access to the Microsoft Office tools you know”. SalesForce, Oracle, IBM, SAP, HP, Amazon and most major IT companies are making strong service offerings built on a web-based model.

We think of Cloud as more of a business model than as a specific technology. If I buy “Cloud”, I probably want to:

  • Stop dealing with infrastructure and focus on my unique, core business.
  • Scale gradually rather than in “2 expensive steps forward, 1 disappointing step backward” leaps.
  • Let my big apps communicate with each other using secure, open standards and APIs.
  • Take advantage of new capabilities and features as they appear, without having to plan costly rebuilds and upgrades.

Most customers who move to CMS in the Cloud are trying to get more agility in their web practice, start focusing on achieving some real goals, and stop managing their web channel applications.

At the same time that Cloud is pulling web delivery out of the internal data centre, many organizations are dealing with a bring your own device strategy, with new and unmanaged devices needing access to information and systems which were traditionally only available on the local network.

These trends are creating new issues for CMS platforms:

  • Application scalability: Many CMS applications are not built to operate in a cloud architecture. Operating without fixed physical memory, storage, or a traditional database may not be possible without substantial redevelopment. This development is costly and can be high risk. Many vendors have their licensing model tied to processors or physical infrastructure, so their motivation to make that investment is low.
  • Security model: Many CMS applications provide a variety of models for managing user authentication, both for internal content authors as well as external content consumers. ADS and LDAP have been the standard methods for authenticating internal users. Cloud authentication mechanisms add complexity to this architecture, however, especially when users are coming from outside the company network.
  • Content permissioning: The CMS is becoming the integration point for web content, internal enterprise applications, as well as third-party cloud platforms (i.e. SalesForce). Clear identity management and permissions frameworks are an important UX problem for CMS designers, because making mistakes with identity carries an enormous reputational risk.

Every CMS vendor will have to deal with Cloud scalability eventually, but it will always get put on the back burner in favour of more sales-oriented capabilities. Remember, customers are still shopping for that next “killer feature”.

6. Wrestle Search back from Google

2013: There’s Google, and then there are organizations that haven’t shelled out for a Google Search Appliance yet. Google has become the “put a bird on it” of enterprise search, and other vendors can’t keep up with Google’s rate of innovation and commitment to user experience.

Maybe we just give in to the awesomeness that is a Google Search Appliance? Our team prefers GSA for ease of extension, capabilities, and time to market. If budget wasn’t an issue, I’m sure we’d ask every customer to buy a small rack of GSAs and be done with it.

Aside from the cost of deploying GSA in high availability web environments (you can always use Google Site Search for a lower cost solution), the failure of CMS vendors to keep up with Google on search is going to cause other problems for the customer.

Extranets and intranets cause more problems. While private search collections are straightforward with GSA, it takes content permission management (i.e. who gets to see what) out of the CMS and puts it into an interface that the content owner can’t normally access. If content is private, you don’t want it showing up in search results. When content has complex permissions, search results should be the CMS’ job.

Google does what 90% of end users want better than anyone, but some advanced capabilities (faceted search, case management, and discovery) require significant custom development, or just aren’t possible at all. By surrendering the 90% to Google, CMS vendors make it difficult to create a unified experience when more complex tools are needed.

7. Tell me who I am

Identity management (IM) is becoming fundamental to a successful web channel. For most organizations considering a CMS change, the ability to personalize, localize, and target content is a key consideration. A CMS needs to deliver content in an identity-aware architecture, but also give content authors usable interfaces to create and manage identity-aware content.

Identity management causes a lot of complexity in a CMS deployment:

  1. CMS users need access to the system, whether they’re on the internal network or not. The CMS needs to connect to Active Directory or LDAP directory services to understand who can access and edit content.
  2. The CMS needs to deliver content in an identity-aware fashion, support usable security patterns, and be compatible with single sign on platforms.
  3. Security is important. But it’s not always clear how to manage security – through internal directory services, in an identity management platform or as a registered user in the CMS platform.
  4. External customers can sometimes create an account to store preferences and history, or to participate in online communities. This user account may be managed by the CMS platform. Most CMS platforms that provide this capability do not also have mechanisms to reconcile identity information stored in Customer Relationship Management (CRM) systems with internal (LDAP) users and with single sign on applications.
  5. Site administrators may want to generate analytics on user behavior, but Google does not permit personally identifiable information to be passed to their Analytics platform and few CMS vendors have made the investment into analytics to provide this type of data.
  6. It’s a common pattern to see external social credentials being used to enable sign-in. While this gives Facebook a lot of valuable information, it doesn’t help site owners at all.

Identity Management is a practice that most organizations struggle with. CMSs should be able to help them – but only with better tools and infrastructure.

It’s never what or how, it’s always who

A good friend of mine has a saying: “It’s never what or how, it’s always who”. This should be the mantra for CMS vendors and communities. CMS platforms need to benefit people.

A CMS isn’t a solution, it’s a tool. So many CMS platforms use value propositions like:

  • “An intuitive design interface encourages business users to create a web presence based on the principles of responsive design to deliver optimized, non-disruptive experiences.” [OpenText]
  • “...provides unparalleled usability that removes the roadblocks associated with getting more content online, in more places, faster than ever before.” [Percussion]
  • “...helps you organize and manage the delivery of creative assets and other content...” [Adobe]
  • “...insert YouTube video or Google Maps into the text...” [Kentico]
  • “...even casual users can easily edit content right on the page...” [SiteCore]
  • “Enables marketers and business users to easily create and manage contextually relevant social and interactive online experiences across multiple channels...” [Oracle]
  • “Create and deliver web pages without developer or IT support...” [Ingeniux]
  • “Revolutionize page and site creation by empowering content authors and marketers to manage the entire page creation process in a simple, visual interface.” [Ektron]

It looks like CMS marketers think that the number one reason to buy is to make it easier for non-technical users to shovel content onto the web.

My hope is that within the next two years, the reasons customers go shopping for a new CMS will look like:

  1. “We need to provide data to our content authors about how their content is actually working”
  2. “Our web presence is an integral part of our entire business, and the CMS needs to be able to integrate with our other enterprise tools”
  3. “Our current product can’t facilitate the kind of workflow we need to manage risk”
  4. “We can’t rely on WYSIWYG editors, we need to create a true enterprise content repository and store structured, reusable content”
  5. “We need our entire architecture to move to the cloud, including our CMS”
  6. “Search is too integrated into our content creation process to try to integrate 3rd party search tools”
  7. “Our content management practice relies on knowing who is browsing our site and we need a content creation platform that lets us create content for our actual customers”

The CMS is a tool for humans to create content for other humans, at mass scale. That has always been and continues to be a tough job. The future has to make it easier.