Sites are Hard, part 1


Ben Bose




shame, scandal, own dogfood

Back to Articles
Ben declares...

tl;dr: Ben complains that he had to do some work

Mavens was established in 2009 (it says so on our logo). It's now 2010, and we've just managed to launch a site. We are (allegedly) good at the Internet. So why did it take so long?


Well, one thing that anyone who builds anything can tell you is that building for yourself is tricky. All those objective pieces of analysis - so easy to make for someone else - somehow become nearly impossible to carry out effectively. Oh, and when your shop window is, erm, the shop window, then you have to make sure that it's a damn nice window. Add in a touch of being a neurotic perfectionist, and no single owner for the site, and you have what the Urban Dictionary refers to as a Mongolian Clusterfuck.
[Warning: no-one should ever visit the Urban Dictionary - Ed.]


These are all really excuses, though - we do know what we have to do, and how to do it. So, at some point towards the end of 2009, Leon and Ben sat down one evening and implemented the purest form of the agile method. We grabbed a cloud server, installed WordPress, bought a skin, and edited it into something resembling the Mavens' brand. [We since changed the brand, possibly as a direct result - Ed.]

It looked something like this:

Custom Blog Skin by Mavens of London

Early reactions were mixed, apart from the board. They hated it, though they were pretty polite to Ben (he cries easily). So we had another go - this time by getting some design concepts created - basically around getting some brochureware up, since the lack of a site was embarrassing everyone.

It looked like this:

Early Website Designs for Mavens of London

So, yeah. That was way too much pipe.


So Ben finally sat down and did the work he'd have told any client to do (or at least pay him to do) right at the start of the process. He:

  1. Looked at search terms around online marketing;
  2. Looked at the sites currently ranking around the terms, and created a features analysis;
  3. Generated an early content plan matched against the search terms;
  4. Added a webcomic to the plan;
  5. Listed functions against the content requirements, then translated those into plain wireframes;
  6. Matched wireframes to contents to generate an initial sitemap;
  7. Created a list of URLs, titles, and headers for the initial sitemap;
  8. Made some notes on how to get some interesting content once the site was live, and integration with other sources of content/data (mainly LinkedIn, owing to a obsession with APIs at the time); and
  9. Put it all in a document called "Wibble":

Wireframes & Content Plan for Mavens of London

This took him pretty much all of Saturday. He then showed it to Leon, who seemed really confused. Satisfied, he then arranged for a meeting with some people who could make it all look pretty.

Next time: Early designs! Unnecessary revisions! Internal politics! An example pattern library!