How long will it take me to impliment CMS for a client?

July 24, 2009 12:10 PM

How long to deploy CMS?

I have STRONG web design and development skills and I have been asked to give a quote for implementing CMS for a client as part of their rebranding efforts. The catch is that someone else is spearheading the rebranding, which will include the new look for the site -- with input for me.

So hive mind how what is a reasonable estimate for how many hours it will take to develop, deploy and test a CMS (like Drupal) for my client?

For those folks who have done this already, I would appreciate any hindsight advice you might share with me.

Thanks in advance for your opinions, time and advice!

posted by Fuzzy Dog to computers & internet (11 comments total)

It obviously depends on the scale of your project Can you give me an idea of how large the userbase will be, and how much content will be in the CMS?
posted by lazaruslong at 12:19 PM on July 24

Thanks for the quick response, The site has a small user base, 8-10 content creators, but most of the content creation falling to 2-3 people, lots of content - 30 LONG primary pages in 8 categories with lots of sub pages. Let me know if that does not cover your question.
posted by Fuzzy Dog at 12:29 PM on July 24

Will you migrate data to the new site? How much data? From what format?

What features does your client want implemented?

Are you going to train the staff in using Drupal?

Have you made any plans for maintenance and upgrades? Who will make sure that Drupal keeps running when you're gone?
posted by Foci for Analysis at 12:36 PM on July 24

Drupal's pretty fast, I did one as a student project in a matter of weeks. Could take longer if you want to take advantage of more plugins.

If you can get by with something less robust FrogCMS is totally simple and can be done in days.
posted by bitdamaged at 12:36 PM on July 24 [1 favorite]

How long does it take to build a house?
posted by shownomercy at 12:47 PM on July 24

How long is a piece of string?
posted by rokusan at 12:50 PM on July 24 [2 favorites]

The way you're asking this questions suggests that you have no experience with developing for and deploying CMSes. Apart from the fact that you're not delivering the whole thing but are dependent on someone else doing conceptual and visual design, in my opinion neither you nor anybody on MetaFilter is in a position to estimate the work that needs to be done based on that.

Generally I'd say you should not do the job at all because it wouldn't be right to do this for a client, if you have no experience. As a solution however you could tell the client upfront, then ask for their budget and do the job within the budget (if reasonable) and learn working with CMSes while implementing it. Make sure there's enough time for your learning in the schedule.
posted by oxit at 12:58 PM on July 24 [2 favorites]

The only way you are going to be able to answer this question is if you are the project manager. If you are not (and you are not) then you should not be held to answer this question, as it is unanswerable from your current viewpoint.
posted by BrodieShadeTree at 1:04 PM on July 24

You also don't mention what city/country you're talking about. The going rate varies widely.
posted by rokusan at 1:24 PM on July 24

There are many definitions of 'time' and it sounds like you will be exploring many of them during this project. From knowable to unknowable:

1. Time spent by you, in total, banging on the keyboard to get things going.

2. Time spent by client in their rebranding effort.

3. Latency time between #2 and #1.

4. Total time, measured in days, from the start of the project to the end of the project where 'end of project' is hopefully less than 'estimated time until heat death of the universe.'.

I have a feeling #1 is the answer you're most curious about but is also probably the least relevant to the entire endeavor. If you're a good developer, it shouldn't take you more than 20-25 or so continuous hours to bang together a Drupal or Joomla site. That is also the easiest part of the project.

The most difficult is the interaction between you and the rebranding-person. Because they are spending time and money on re-branding, there must have been something fucked up about the previous brand - either that, or they are just wasting some money. Money wasted in this fashion must be very _precisely_ wasted, in such a way that might dictate pixel-by-pixel placement of pointless screen widgets.

After the 'look' has been more or less completed, you get to move on to the 'feel', which is being techsupport for all of the people generating the content. I would examine the format/process they currently use to generate and manage content because the less change for the users, the better. If they're used to saving everything into Wingdings under Microsoft word before uploading to a staging server, then you better replicate the process or streamline it in a fashion that lets them change their process as little as possible.


The biggest area of focus for you on this project should be a clear, contractual understanding of exactly WHAT the client is asking for and where your boundary of responsibility ends. For instance, are you going to be responsible in a couple months when some weird injection attack comes out and their site gets exploited and filled with pornography? Think about every single thing that can wrong to make your life a living hell during and after and get it all in writing.

Do not be afraid to charge too much. Do not be afraid to define and endorse boundaries on your responsibility. Do not be afraid to walk away if the deal is going to screw you.

I don't mean to sound pessimistic, but such projects can be fraught with pain, trauma, and marketing people.
posted by Skrubly at 1:31 PM on July 24 [1 favorite]

I've deployed a much bigger site in 2 months. Everyone had a very clear idea of what they wanted and there was never any placement text, articles, graphics or categories. It already existed in their old system.

On the other hand I'm (still) in the middle of a site deployment for a smaller site than you spec. It has been going on for over 18 months. The client doesn't have a f-ing clue, and the project has gone through three managers in this time, all of whom have tried to change the entire project, and they expect me to take input from everyone at the company. Including the stuff that contradicts the other stuff. About 2/3 of the pages on the site are temporary because no one can get the contributors to contribute.

Hope for the first client but write a contract that will take into account the second. Include account how much you get paid for each milestone, along with spelling out what qualifies as reasonable revisions and the process for change requests. And then put down how much you charge per hour for going beyond that. A good contract took that second client and transformed them from a giant PITA to a cash cow.
posted by Ookseer at 4:57 PM on July 24 [1 favorite]