iMarc.net Redesign: Part 10, Project Retrospective
by Dave Tufts - May 17, 2007 / 1:01pm View more articles
iMarc just redesigned our website. These articles chronicle that process.
« First Article (Part 1)
« Previous Article - Part 9, Launch
This past Monday marked the one week anniversary of our new website. With our lead designer Craig back from his Galapagos excursion, we decided to have a project retrospective.
The general concept of a retrospective, or "post mortem", is to bring everyone involved with the project together after all the work is done. Team members discuss what went well, what could have gone better, and what they learned.
For this retrospective, we got the core team together – the designer, project manager, programmer, copywriter, and information architect. Even though we had input from a much larger group, only the core team attend the retrospective to preserve some semblance of sanity. Fred, a senior developer who was not heavily involved in the programming on this project, led the retrospective as an impartial outsider.
Our retrospective mirrored the following process:
- Leader asks a question (Are you proud of the finished product? What was the most frustrating part of the process? What's your favorite part of the new site?)
- Team members answer.
- The leader takes notes, prompts follow ups, and promotes further discussion on interesting points.
- (repeat steps 1–3 for five or six questions)
- After the retrospective, the leader summarizes everything in a one-page PDF and sends it to everyone.
We learned a lot from the retrospective. Actually, the real learning came during the process of building the website. The retrospective just a acts as a vehicle to identify and archive what was learned.
Highlights from the list of things we learned include:
- Going from comp to cut up site, the comp can be rough. Don't sweat the details, as long as the design concept is strong.
- Grey boxes help bridge the gap between concept and end product.
- Determine stakeholders early on—demand their involvement.
- Design by committee may work if everyone is knowledgeable on the topic, although things get better as the group gets smaller.
- Branding is more than just a logo—we should explore our own branding in much greater depth.
On a more mundane technical note, we learned a lot about dark sites and how light text on dark backgrounds renders differently on Macs and PCs. In our first color scheme, the text was almost white and the background dark grey. On PCs this looked fine, but Mac users reported major readability issues. Our fist instinct was to increase contrast by making the background even darker. This worked on terse text blocks with very short line lengths, but made reading a multi-paragraph blog even more difficult. After some research, we found lightening the background – decreasing the contrast – increased readability across both platforms.
So that's it. Project done. Website launched. Knowledge archived. Everyone happy.
If you want to review every step, here's a quick index of the entire blog series:
- Part 1, Looking Back
- Part 2, The Team
- Part 3, Information Architecture
- Part 4, Same Page
- Part 5, Grey Box Comps
- Part 6, Tools
- Part 7, Comps
- Part 8, Sculpting
- Part 9, Launch
- Part 10, Project Retrospective
The End.
Comments have been turned off on this blog.
Read something more recent.
2 Comments
First though, this is amazing. A fully documented design, development and deployment process. Probably one of the first and most honest I have ever seen. I applaud you for also putting it into a format to which people like myself can comment.
But,
A blooper reel. Notes about the hiccups along the way. It may be just myself, but I have a feeling the process wasn't as smooth as documented.
Of course, you're correct:
- we wish we had more time
- we wish we implemented sIFR
- we should have solicited more direct feedback from the sales and marketing team
- in the past two months the we've probably spend 95 hours arguing about the name of a particular PHP...
- I'm sure others could vent for days about annoying things I did!
But all in all, it went really smoothly.
+ Maybe because of compressed timeline - not enough time to get totally dysfunctional.
+ Maybe because we really didn't change THAT much - compare this site to the last. There are lots of small changes, but nothing major.
+ Maybe because we kept things so open and honest.
+ Maybe just because we're so good!