May 2013
« Jun    

How to Avoid CMS Selection Confuction

The backstage manager was pacing all around by his chair
“There’s something funny going on” he said ” I can just feel it in the air”

Let me tell you a secret. I probably shouldn’t, but I’ve been reading a few more “How To Select A CMS” posts recently and, truth be told, they’re the same as they were when I stopped reading them way back in 2011. Exactly. The. Same. Which is a bit depressing really.

So this is just between you and me: The features of the CMS don’t matter. At least not in the long term compared to the big things that’ll fuck you over. You know that big RFP feature matrix? Horseshit. The lovely demos in the vendor Dog And Pony shows? Yawn. Interviewing all your stakeholders? Ho hum – they don’t really have a clue.

Sure, you might have a decent idea of the features you need today. Maybe even in six months? But you don’t have a scooby what’ll happen next year. You could get broadsided by your vendor going to hell. Or that new CIO who wants to (de)centralise. Or that company merger. Or your department mandating Technology X or Open Source or Some Insane Security Policy. Or the thing that doesn’t exist yet that’s going to disrupt your ass back to the stone age.

Here is what matters. Your content needs to be able to outlive the technology that houses it. This is the most important question you should ask in any selection process:

Question 1: “If we select you, and it all goes tits up, how would we get our content out of your system and into something else?”

If it is painless to get your content out in a sensible form, you can’t go too badly wrong on your product selection. This is especially true if you don’t need to shell out a whole lot of cash up front for the software, but you are paying-as-you-go.

So that’s the software problem solved. But, as we all know, it is actually the services where you’re spending a whole lot of cash. So you’ve got your integrator estimate of a few tens or hundreds of man days. That’s a lot of bodies scuttling around on your project doing a whole lot of very important things. So take a step back and ask yourself the next most important question.

Question 2: “What the fuck are all those people actually doing, and what part of that work could we keep if we moved to another system?”

Some of the work is independent of (and should probably predate) the technology selection. Strategy, design, information architecture, content type modelling, business process modelling and all that good stuff. But on the technology side, there are some pieces that take a fair bit of time and should result in a portable asset. At the very least, all work creating the actual HTML should be independent of the system. If you’re using a standard templating language like XSLT (bite me, @pmonks) or FreeMarker or mustache.js, parts of your templating layer could migrate. And of course any hardcore business logic should be exposed by an API that could outlive your CMS. But if all these busy systems integration bees are plodding away on tech that you need to chuck should you change system, you’re probably doing something wrong.

So it is as simple as that. Read your Darwin again. Change happens. And it will keep happening at an ever increasing rate. Pick something that can change with it. Make portable content and portable technical assets. Don’t make monoliths.

  • Facebook
  • Google Bookmarks
  • Digg
  • LinkedIn
  • StumbleUpon
  • Technorati

14 comments to How to Avoid CMS Selection Confuction

  • K

    Thanks for good advice: focus on the data, and getting it back/out if necessary. Came across this post while researching the state-of-the-art in CMSes, which as you note, seems uninspiring at best. Any future-CMS software, people, or ideas out there which you think are worthwhile?

  • a

    Hey there, You’ve done a great job. I’ll certainly digg it and personally recommend to my friends.
    I am confident they will be benefited from this website.

  • You share interesting things here. I think that your page can go viral easily, but you must
    give it initial boost and i know how to do it, just type in google for – wcnu traffic increase

  • You are so right, those two questions are important. Also I would say don’t write a RFP (request for proposal) which looks like a list of requested features. Instead write 5 to 10 user stories and let the vendors present how they would solve the hurdles in the stories. This gives you a way more realistic impression of what the system can do for you.

  • It’s genuinely very difficult in this full of activity life to listen news on TV, thus I just use web
    for that reason, and take the hottest news.

  • 日本人気スーパーコピーブランド通販専門店!
    ルイ ヴィトン、シャネル、エルメスなど有名
    ◇各種各様の最高品質スーパーコピー腕時計、ブランドコピー腕時計 激安
    ロレックス レディース シンプル yumi

  • 皆様の絶好評判を偽物ブランドコピー 時計人気専門 ショップです
    ロレックスブランドコピーなど世界中有名なブランドコピー 時計を格安で通販しております。
    N級品ブランドコピー 時計は 業界で最高な品質に挑戦します。!

  • 日本的な人気と信頼を得ています

  • 画像通りの状態のお品でした。商品の説明も細やかで正直になされていると思います。梱包もしっかりされており、とても信頼できるショップだと感じました。
    【送料無料】グッチ トートバックをセール価格で販売中♪グッチ トートバック GG 130736 ゴールド キャンバス レザー 新品 ホーボー ダブルG GUCCI
    画像どおりのきれいなお品でした ありがとうございました

  • Нужно существовать в хорошо обмишуленном жилье. Удобные свойства трикотажных материалов с микрофибры В наши время экофизиология одежи, природоохранное действия, удобство и облик существования начали значимой составляющей нынешного изготовления материалов. Одеяние, что зажигается, одеяние с стресса, одеяние с блокаторами ультрафиолетовых проблесков и одеяние с кондюком … все без исключения лучший текстиль данное никак не вымышленные нюансы, а данные. В 1 середине ДВАДЦАТЫЙ-спорт столетия интенсивный изнашивание никак не разрабатывался регулярно, таким образом равно как существовали легкодоступны меньше оптимальные естественные и искусственного происхождения волокна. Однако в заключительное период все без исключения данное сделалось допустимым вследствие микроволокнам.

Leave a Reply to Charlesunums




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>