Buyers are normally fairly smart, but tend to forget quite a few things. The web content management industry has now been with us for a bit more than a decade. Today trade press and analyst firms alike like to call the market mature and consolidated. In fact, buyers are still figuring out how to obtain the value they expected.
This year marks my 10th year in the amazing web content management industry and I’ve witnessed numerous things and also learned a few lessons. Many of those apply to other industries as well. Despite being valid lessons, they are often forgotten in the web content management arena. I recently poked my network, including the members in our J. Boye groups about their most significant lessons learned and received some good and often forgotten points such as:
- Expectations, expectations, expectations
- Never underestimate the content migration phase
- Keep usability for the editors in mind when building and configuring the system
- Content should never be managed by committee
- Do not treat content as code or code as content. They are different beasts with different life cycles
- A contract is only as good as your ability to enforce it
- If you don’t consider accessibility from the beginning of your project you won’t create an accessible website
This is all very good advice. However, I am increasingly being reminded of some inconvenient truths in this young and rapidly changing marketplace, e.g.:
- It’s the crew and not the tool – forget about finding the best CMS, but do work hard to find the best implementation crew
- The real horizon for most systems is no more than 3 years – on average organisations change their CMS every 3 years
- Standards are nice, but elusive – we are really lacking standards and while eg. CMIS is an interesting initiative it is still a few years into the future
I did a joint presentation with Tony Byrne from CMS Watch on this back in 2006 and still many -if not most- buyers have a hard time accepting these truths.You certainly can’t expect vendors to suddenly begin promoting the fact that you’ll need another system every 3 years. I’m optimistic about the market but a few things need to change.
Which inconvenient truths can you share?
Jon Marks April 8th, 2009 8:41
Thanks for another interesting article.
I agree completely with your 3 year CMS life-expectancy. I sometimes ask clients how long they plan to use the new system they’re about to implement but, when I do, the answer is normally 5 to 10 years. A 3 year lifespan makes the numbers in the business plans look a little shaky sometimes. For me, the biggest take away from this to tell the client that an extremely important requirement is the ability to easily export all content into a useful format *if* (not when, which scares them) there is a reason to change systems earlier than planned.
I’d also like to make one comment on your point “It’s the crew and not the tool”. The more I play this game, the more I believe that almost any decent CMS can fulfill the requirements of almost any content driven system. Of course there are exceptions, but I also believe that the amount of customisation required is dependent on the skill and tools at “the crew’s” disposal. I believe my crew is excellent, but if we had to implement a new CMS for the first time, we’d certainly screw a few things up as we uncover the quirks of the new system. And they all have quirks! In my experience, getting the assistance of the vendor’s professional services does help, but certainly doesn’t eliminate the problem. I’d like to rephrase your inconvenient truth to “Find the best implementation crew, and trust them to choose the CMS that is best for everyone.” Forget about the big evaluation matrices which add hardly any value at all to the selection process. Everyone ticks everything these days anyway.
Jevs April 8th, 2009 8:41
content….. content content content….. content content content content content content
1) The “launch date” you (the client) are talking about, and the launch date we (the developers) are talking about are likely to be two different “launch dates”. I probably mean, the day that you can start putting in final approved copy and the preparing to spend a few weeks/months tweaking it to its optimum. YOU probably mean the date the site goes live to the world – DEFINE THESE CLEARLY EARLY ON!
2) Any website (and ergo any CMS) lives and dies by the content – plan early and plan well for content production. DONT try and find your old content into the new site.
3) Training and Documentation are important as well as regular staff usage. Have a CMS team at the clients office who work with the system. Yes, you THINK that the CMS means everyone in the office/company can update the website from their desk – but how often does that really happen. If it isnt happening on a daily (at longest – weekly) basis, then youd be better off having a dedicated CMS editor resource.
Michael Marth April 9th, 2009 8:41
Janus,
I find this data point “3 years” interesting and would like to get a bit anal about it
First, I assume this is the arithmetic mean[1]. This number can be very misleading for particular distributions: If 50% of your sample changes the CMS after 1 year and the other after 5 years you would still get an average of 3 years, but would draw very different conclusions from it.
Also, for very skewed distributions the median[2] might be a better number to characterize these distributions (e.g. if there are very few CMS users that keep the system for a very long time and many users that keep it for a short time)
In summary: I would love to see the data
Cheers
Michael
[1] http://en.wikipedia.org/wiki/Average#Arithmetic_mean
[2] http://en.wikipedia.org/wiki/Median
Janus Boye April 9th, 2009 8:41
Hey Michael,
Thanks for contributing to the discussion. As an employee of Day, perhaps you could contribute with some data? What’s the average number of years that a customer uses your system?
I don’t have enough data on my end to be scientific about it, but organizations tend to change CMS as often as many people change their cars, circa every 3 years.
Normally organizations are not too happy to change the CMS. Often it is driven by external forces, such as a major, time-consuming and expensive upgrade, a bankrupt vendor, impossible to find competent help or that the CMS is no longer maintained (e.g. due to merger). Sometimes internal forces are also at play, such as changed requirements, new management or restructuring.
Sometimes the chosen CMS was simply the wrong one from the beginning and that usually takes some time to digest internally, before they are ready to accept it and select a new one.
I wish it was different, but as a customer there is really little you can do here, other than as Jon suggested, make it as easy and cheap as possible to migrate to the next one.
Michael Marth April 9th, 2009 8:41
Hi Janus,
unfortunately I do not have any data that I could make public, but I can assure you that we have a lot of customers that have been with us for a long time (and are happily upgrading, i.e. not just stuck on an old version).
However, coming back to the hard data: maybe one number like “3 years” over the whole industry is a flawed approach anyway. It might be more realistic to look at each vendor’s data separately. One data point that is publicly available is the recurring license revenue and the new license revenue. From them one could probably make a guesstimate about the average customer lifetime per vendor.
Cheers
Michael
Janus Boye April 10th, 2009 8:41
Hi Michael,
I would not be surprised if the data showed the customers of Day stayed with CQ for longer than 3 years. It would be great if Day decided to publicize such numbers.
In many cases where a customer has changed their CMS after 3 years or less, it is because their vendor decided to totally relaunch their system (e.g. Microsoft CMS 2002 -> SharePoint 2007) or when they have used a very local vendor and started to find it too limiting. In other cases the customer have found out that they are the only customer in the country/region and they decide that it is too expensive and difficult to get help. I guess, that if I was a Swiss customer, then it would be easy to find some competent CQ developers, but how about here in Denmark or in Kansas? Upgrade cost is often a driver to change, but this industry has also seen its fair share or mergers and acquisitions where not all products have survived.
Cheers,
Michael Marth April 14th, 2009 8:41
Hi Janus,
(slightly belated due to Easter hols
)
what you say makes a lot of sense to me and reflects my own experience: the reasons and circumstances for switching products vary quite a bit. I guess that’s why I felt a bit uneasy about summarizing them in one number only. But I think we are are on the same page on this topic.
Re the numbers you asked for: I like the idea
can’t promise anything, though.
Cheers
Michael
Michael Marth April 16th, 2009 8:41
OK, I got me the data and did the math…
see here: http://dev.day.com/microsling/content/blogs/main/cmslifetime.html
(spoiler: for Day it’s at least 6.7 years)
J. Boye » Blog Archive » WCM instructions from Philadelphia May 15th, 2009 8:41
[...] the case studies, I did my talk on inconvenient truths based on my 10 years in the industry. The track concluded without slides in a session where three very experienced CMS consultants [...]
Boris Kraft March 3rd, 2011 8:41
Janus
You seem to imply that the CMS has no or little impact on the success of your project, only the implementation crew does. I think this is not what you meant to imply, but I want to share my thoughts anyways.
I know for a fact from speaking with implementation partners that there are systems out there that make it nearly impossible to deliver a project on time and budget; specifically there are vendors who are great at selling something and making their implementation partners bleed when it comes to implementation. I’d love to see statistics on system integrator’s staff churn rate related to CMS.
From another perspective, if the CMS would really be that irrelevant, why should I (as a vendor) invest so much in improving it. (come to think of it, some vendors have certainly come to the conclusion its not worth the investment
).
So, I agree that 1) technology alone won’t make you successful, that 2) if you want to get a job done right, you need to find the right people to do it (nothing new here), and 3) that in addition the internal buy-in and processes to roll out a CMS can make or break your project. But if you are smart enough to get 2) and 3) right, you shouldn’t have a problem picking the right CMS for your enterprise as well.
Regards
Boris