6 practical lessons from our new site launch
- Posted by Nigel Edelshain
- On December 21, 2015
- Websites

We relaunched this website a few months ago and we learned plenty by doing that. I’m pretty satisfied with the way the site ended up but getting there was no cakewalk. Even though I’ve been involved in many website projects before rebuilding our own sites always seems to have its own nuances.
This post lays out six lessons learned from this project and other previous website builds. My hope is one of them will save you a lot of time, money or energy.
1. A lot of this is about people
There’s something about websites. Something that may not be obvious if you’ve not been involved in “redesigning” one before.
Websites are like lightning rods. Most company websites cross many departments. Each department typically has some say in how the site looks. This means building a corporate website involves a lot of “cooks”.
It’s not easy to build consensus. It’s even harder when graphics are involved. The way something looks is very subjective. Everyone has their own view of “good” graphics. We website developers need a few tricks to get to consensus on the graphics for a site.
One of these is a “divide and conquer” strategy. We split out the logic of the website from the graphics. We describe the logic of the site with a sitemap and wireframes. Neither of these contain graphics so it’s impossible to object to either based on the colors and fonts used etc. Then the graphics are presented in the form of “drawings”—non-functional web page designs “drawn” using InDesign or Photoshop.
Keeping these two decisions separate increases the odds that we can get both the logic and design of the site approved. It’s still not easy with lots of people involved but it greatly increases the chances.
2. Mobile first is a new habit
I thought I had a good handle on this “mobile first” concept going into this website build. I’ve dealt with mobile as part of our web development projects several times now but this time I realized we needed to go “all in”.
Whenever we reached a point where we could make a web page look better on a phone or on a desktop, we chose the phone. This made some of web pages look less “glamorous” on the desktop.
There’s some reconditioning to do here for veteran web developers. If you’ve spent many years aiming to make your site look great on a desktop/laptop then you really have to pay attention to the tradeoffs needed to succeed on that tiny phone screen.
When you’re designing for mobile you’re not just designing for two screens. You’ve got phone, tablet and desktop and those tablets and phones are used in both portrait and landscape mode. Now you’re dealing with a least five ways to display your site.
During our website build we found that menus did not always behave the way we anticipated once we looked at them in a particular permutation of device and orientation. “Hmm, that menu looks odd on an iPad in landscape mode”.
Don’t underestimate the time it will take to test on all these combinations of device and orientation.
3. Don’t forget the SEO!
Google drives a lot of traffic to most websites. Getting a new website to look good takes so much effort that performing any SEO work at launch can seem just too hard.
But you have to do the basics: Come up with web page titles and alt image text that contain your keywords. Put in the meta descriptions for all your key pages so people will click on your search results.
4. Change is certain. Be Prepared.
One thing we can guarantee is change! You need to use a content management system for your website.
We made it a full 48 hours before we needed to add a new web page that was not there at launch. 72 hours in we need a new landing page as well. It’s got to be easy for you to deal with these requests. The only way is to build your site on a CMS (like WordPress.)
5. Can you live with your theme?
Initially all you want is a theme that looks good. But you have to live with your theme for the next few years.
Once we were deep into working with our chosen theme, we found several areas that were tricky to manipulate. Had we known about these areas upfront we might have considered another theme. We’ve worked around those “bumps in the road” now but we’re also testing other themes for our future projects.
Just looking good is not enough for a theme.
6. Blogs don’t import themselves
Our blog used to live on a different CMS. We needed to import all our posts to WordPress. We thought this would be simple but like so much in technology “simple” turned out to be a relative concept.
We did import all the text from our previous blogs posts but a lot of the formatting had to be cleaned up manually.
Our featured post images did not import. We also realized the aspect ratio (height to width) of our new blog images was significantly different to our old blog images. The upshot was that some of our old photos did not work at all on the new blog (cut off heads etc.) So we had to find brand new photos for certain old posts anyway.
Given these issues, we had to proof all our old posts (around 200 of them) so we kept our intern busy reading old blog posts for quite some time.
2 Comments