Since returning from my secondment to the Govt.nz project, I’ve been using what I learned to improve the search engine optimisation (SEO) of the Work and Income website.
The Work and Income brandI thought that most Kiwis would navigate straight to the website because it’s so well known. I didn’t expect high numbers for people coming from search engines.
Imagine my surprise that since January 2014 around 72% of external sessions originated at organic search (eg Google, Bing etc). In September 2014 the figure was around 75%. Over half of that traffic just searched on the Work and Income name but the rest were searching for specific content.
Where to startI started with research. I had to do some serious number crunching to get the full picture on how our visitors use search and what they’re searching for.
Luckily I have access to a number of Google tools to drill down into our users’ search behaviour. Google Analytics is a must, but I found that Google Webmaster and Google Trends were invaluable to me.
When I started analysing the search data I found that some of our content was not ranking well in popular searches. Even though a lot of content ranked well on Google, some content (especially around job search) wasn’t even on the first page of search results.
Doing the workAfter gathering all my evidence I was able to produce recommendations and get sign-off to start improving the site.
I came up with a work plan and identified the pages that I was going to work on first. I looked at the metadata and page structures of priority pages that were not ranking well in search. Then I:
- added keywords or phrases to summaries
- made sure page titles or headings matched frequently-used search terms
- added pages that were “missing” from the site (eg a highly searched term with no actual content on our site that matched it)
- ensured pages had keywords that were relevant to the page of content.
Telling Google about the siteAfter I did all this work I knew the site needed to be “crawled” again, so that Google could see the changes I’d made. I could do this through Google Webmaster.
Work and Income’s website was all set up in Webmaster but there was no sitemap loaded.
Sitemaps are important as they let Google and other search engines crawl your site more intelligently.
There are lots of ways to generate a sitemap. Some Content Management Systems (CMS) can do it automatically, but I didn’t have the option. I used one of the free tools available online that let me quickly create a site map in XML format.
That covers Google, but what about other search engines? Over 10% of Work and Income search traffic comes from them. I thought I’d have to set up Webmaster accounts for all of them as well to ensure they were crawled correctly. Then I found a handy little bit of code to put in my robots.txt file:
This tells other crawlers were my sitemap is, allowing them to index the site correctly.
Correcting errorsAnother great feature of Google Webmaster is that it can also identify ways to improve a site’s SEO. It gave me lots of tips that I used to improve the site. I was able to eliminate duplicate page titles and summaries from the site, which also helps.
Will it work?I don’t know. But I’m already seeing some content going up the Google rankings less than 2 weeks after I started making changes.
One of my pages was previously ranked at 42 in Google 3 weeks ago, and it’s now ranked number 1.
This gave me more evidence that what I’m doing is working.
What’s nextI’ve just got started on this work and there’s so much more to be done to improve the site’s SEO. I’m going for iterative improvement using evidence.
I’d love to hear feedback on what you’re doing to improve the search rankings of your site.