IA for the Small Site

a walk down memory lane

[grandpa]

Information Architecture for the Small Site [webmasterworld.com]

The linked thread originally appeared in April and, as is typical of WebmasterWorld, it is a keeper. I'm in the very first stages of a new site and a new adventure, contemplation and consideration. One might say I haven't done a damn thing yet :) one would be mistaken

A few questions came to mind as I was reading the posts. One I would really like to know is, among those who made changes to their sites based on the given information, did you see any measurable changes with your site? I do have one site now where the Rule of 7 could be applied. However, it is an established site with lots of repeat business and I feel that a change might alienate a few people. Folks don't like things to change once they have achieved a comfort level.

Additionally, did those changes affect your SE positions? I'm thinking only from the perspective of an established site.

For a new site; the one I built in December uses the inverted L method of navigation, and it seems to have a positive effect on the site. Example: my PR is more equally shared than with a site that uses a hierichal menu. There is more grouping of the main categories on this site also, with individual options branching across the top of the page. And the site utilities are all presented with a few links on the bottom of each page. However, I do think some minor changes will be made based upon what I've read to date. No, I didn't get it right the first time... close, but no cigar.

[tedster]

About the "Rule of 7" I recently appplied it, with a gentle touch, to a client site that had 16 menu items listed vertically. I broke them visually into four groups, separated by an icon.

Group 1 - 3 links (italic font)
[icon]
Group 2 - 5 links (regular font)
[icon]
Group 3 - 5 links (regular font)
[icon]
Group 4 - 3 links (italic font)

My favorite stickiness metric, page-views per unique, is up over 60%. This is a lead generation site, and it appears that leads are taking a similar jump, although it is a bit early (less than a week so far)

[grandpa]

On the new site that I mentioned that metric is horrible. In on one page, click once, and gone again. I've got to have another look at the 'on page' factors and try to figure out why I don't have stickiness there. Color scheme, content, product offerings, search terms used? From the IA perspective with regards to navigation I think the site is good... maybe too good.

[tedster]

If you look at the search terms that are bringing traffic, and then look at the labels you are using in the menu - is there a nice "resonance", so that the keyword chosen in search seems like it would find value in clicking on further links?

And of course, there are always the possible server or browser technical problems. I recently fixed a newly problematic page -- I had done an edit and only checked in Firefox afterward. But the page was now blowing up in Explorer and until I saw the stats I didn't realize it. Sloppy, sloppy, sloppy!

[grandpa]

I tend to believe the keywords are matched to the pages, so there is resonance... like I understand what I just said :)

I tend to believe my page-views-per unique are more a reflection of the visual aspects of the pages. I tried to be minimalistic and functional when I designed the site. Descriptions are bare-bones, colors are probably awful, and some of the navigation isn't that intuitive.

So I'm working on 2 of those right now, nagivation is being revamped a bit, based on the discussions of the original thread. The visual aspects of the site are being changed, maybe it will look a bit more professional although still minimalistic. The content will simply have to be started over.

I guess I'll have to go hunt down some of the posters from the original thread to get more answers about what affect, if any, did their changes produce.

[grandpa]

It's early yet, but after making some navigational modifications and de-hancing the look of the site, the page-view-per-unique have doubled over the last 3 days.

Another interesting aspect.. after learning how to write valid pages I generally stopped checking them with the validator. Big Mistake! There wasn't anything that a visitor might notice, unless they examined my source.
Lesson learned: if you make a page template, validate it before generating a bunch of pages with it.

[tedster]

page-view-per-unique have doubled over the last 3 days.

That's what I'm talking about - MAJOR jumps. Damn the eye candy folks, slick ain't sticky. Focus your efforts on IA, intuitive menu labels and such. Make the graphic elements simple but elegant -- and that can be harder than making them showy.

Grandpa, I'm so glad to hear about your preliminary results.

[ergophobe]

de-hancing the look of the site

I love the term! I'm curious, though, if you could give some specifics. Do you just mean making it less graphic intensive and pretty but more usable? Does it have fewer images, or just cleaner images? Short labels or clearer labels?

[grandpa]

The site logo was made over - looks more like it was done by an adult and not a 6 yr old with a full box of crayons. The page background color was dumped for white, and the navigation was cleaned up some. After changing the look and feel, I added some aff banners, so the image count actually went up. The banners (for the most part) were selected to blend in with the site color theme.

All in all, I think it looks a bit more professional. Keep in mind though, I was quite happy with the first incarnation when I created the site.

And, still watching the stats, the unique page count is still up. Now if could generate some real traffic...

[ergophobe]

looks more like it was done by an adult and not a 6 yr old with a full box of crayons

Wait, are we talking about my logos? ;-)

I just today built a template for a new site and this thread and its progenitor have been running through my mind and I think (hope!) the new site will be better for it. It's fairly clean, keeping in mind the rule of seven and such.

Tom