Measuring the results when you update your website

It’s really important to make changes to your website on a regular basis to try and improve the usability, give users what they are looking for and of course increase conversion rates.

But how do you know if what you are doing is working?

I follow a few simple rules to make sure that the changes I am makig are having positive effect on my sites.

1) Before making any changes, I compare the landing page on my site to my two main competitors and ask myself ‘what can I do, to make my page better than theirs?

Can I make the layout easier to navigate or the content more in depth?

Point to remember “Google wants to deliver the most relevant results to its users, so creating a well laid out informative site is a ‘must’.

2) I have a definite goal for the changes I’m making to my site, are the changes being made to directly impact conversion rates or are they being made to improve usability and reduce friction (the time wasted when your users are unsure of what to do next) these kind of changes have an indirect impact on conversion rates by increasing the trust factor of your website. Usually, the longer people stay on the site and the more interactions they take, the better.

Example – The change I am making today is a layout change to make navigation easier, ensure that users understand ‘what to do next’ and make it easier for them to navigate to a page which can answer their questions in detail.

Point to remember – Make sure that you add an annotation to Google Analytics, so that you remember when you made the change.



Increase in traffic?
First, let me tell you a couple of things – HITS are how idiots track success, it’s great to get traffic to your site and it’s definitely a metric that you should be reporting on, but it is far from the ‘be all and end all’

Length of time on page
Again, this is a factor which many online marketers report on without really understanding what it means, people automatically think that spending longer on the page is a good thing. But not necessarily, in this instance our goal is to make navigation to deeper pages easier, so reducing time spent on the page we are optimising would be a positive factor.

Let’s think about this from a slightly different angle, Mr Client Persona A, has minimal knowledge of our product. He knows a little bit about it and he’s pretty sure that he needs it but that’s about as far as it goes.

Example – if someone searches for lawyer in Bangkok, they know that they need a lawyer to help them fight their ex partner for custody of their child. But many people still don’t type specific queries into Google, instead they type ‘law firms in Bangkok’. They click on your search result and land on a page which immediately enables them to see the service they are looking for and click through to a more detailed internal page.
If they are more tech savvy and they use a search query like ‘custody lawyer bangkok’ then if your site is optimised correctly they should land on the specific child custody page.

So, what would our goal be regarding ‘time on page’? That’s right, we would see a reduction in time on page but an increase in flow to the inner pages as a positive metric.

How about bounce rate?

Bounce rate is an odd metric, and can be misinterpreted easily. The best way to deal with bounce rate is to decide what is classed as a ‘failure’ for your page? If someone leaves straight away and clicks on another search result that is almost always a failure but if they stay on your website for long enough to absorb the information they require then this shouldn’t be classed as a bounce.

You can adjust your bounce rate by adding this snippet of code just before the closing script tag in your Analytics tracking code –

setTimeout(“ga(‘send’,’event’,’adjusted bounce rate’,’page visit 15 seconds or more’)”,15000);

You can set the time allowance to how ever long is appropriate for your site, 15 seconds is usually a good starting point if you are not yet sure of the best time for your site.

I usually advise people to make a decision based on their average time on page , if you want to get really advanced then you can set different adjusted bounce rates for different pages on your site.

If you are changing the layout of your page to improve usability and behaviour flow then you would be looking for a considerable reduction in your bounce rate. In this case, the bounce rate on the Bangkok law firm site reduced by 15% in the four weeks following the change.

Drilling down deeper

Don’t just take your results on face value, make sure to drill down deeper and look at the results from different traffic sources and different locations by setting your secondary dimension. In normal circumstances, organic traffic tends to have the highest conversion rates and the lowest bounce rates. If that is not the case on your site then you may need to do an SEO audit to check for problems.

So, after making any changes to your site, wait for 7 days and then use Google Analytics to check if your changes are having the desired affect.

Continue to monitor for four to eight weeks after the change and make comparisons to the following time periods

  • just before the change,
  • a few months before
  • the previous year

Good luck, and if you need any help with any of the items discussed in this article, please contact me using the form below and I’ll see what I can do to help.

Is it still worth creating a Google My Business Account?

With all the controversy around Google Local lately and the fact that there are now just three local listings shown instead of 7 made me wonder if it is still worth creating a Google Local listing.

The PIN cards seem to be taking forever to arrive these days and on more than one occasion I’ve had map locations randomnly change to a different location sending customers to the wrong place amongst other problems.

It’s getting harder and harder to encourage customers to leave reviews on your Google listing because it’s not a straight forward process unlike Facebook where most people are more than happy to leave a review without even being asked.

However, the signals that a Google business listing can provide are still valuable for local SEO, so setup your page, get a handful of local citations using exactly the same address, make sure to get at least 5 reviews and embed your listing onto your website.

Although Google My Business probably isn’t going to be a quick route to the top as it once was, it’s still worth the effort!

Data Mining for Bad Back Links

There are some jobs that have to be done, no matter how boring and time consuming they are and data mining is one of them.

There is plenty of software out there which can help you, but nothing is ever as good as your own instinct and common sense.

However, it’s not always as boring as it seems, finding the kind of links that are pointing to a specific site can get quite interesting particularly if you discover a certain pattern.

But it does get disheartening when you are following all the rules and creating great content, building relationships with other authoritative sites in your niche etc and you see other people building nasty, spammy links and ranking above you – but for how much longer you ask? Surely sooner or later Google will catch up with them!

It’s amazing the tricks that people are still getting up to, links in the page title, keyword stuffing, crappy blogs, links in forum comments that specifically tell you not to add a link!!!

I always say that you should take every negative experience and try to make something positive from it, so I can use this data to warn people to check their backlink profile on a regular basis. Particularly if you have been paying ‘best seo company in world’, (thanks Carl Heaton for the voice that is now stuck in my head) to build links for you.

Every three months, download your latest links from Webmaster Tools and take a look at who is linking to your content. If you don’t have access to Webmaster Tools you can use tools such as SEO Spyglass or Opensite Explorer by Moz.

The quickest way to take a sneaky peak at your backlinks is by typing links: into your browser and viewing the results – you may be surprised!

Make sure that you do replace domain name with your own actual domain name though!