It is only when you get to the point of completely rebuilding your website, as we are at the moment here at Channel, that you remember just how much content you have built up on it over the years. For many people the migration of their website is just a matter of getting a new design and structure that they like, but for an SEO you impose on yourself the additional headache of not wanting to lose the precious rankings that have taken you so long to build up.
There are some quite terrifying stories of companies that have gone to the wall, or nearly done so, because of catastrophic loss of rankings when their new website went live. I'll need to blog about that - if I can do so without pointing the finger unnecessarily!
Now, as well as changing our website structure and design, we have also decided to hit the nuclear button and change our domain as well. So we know that our rankings will drop to absolute zero, and that it will be necessary to work them all the way back up again. Other SEOs will, I hope, feel for me here, but we decided that the name change was essential and the rankings loss was a price we are prepared to pay.
This is something that happens to our clients quite often, and it is our responsibility to wok their sites back up from zero to where they were - and then of course onwards and upwards! But we are often asked "How long will that take?". Like most honest SEO companies, we have to say that "We don't know", because Google can and do regularly change the rules. This is not something we are in control of, therefore, we are not really in a position to give any guarantees. But what we can do, of course, is give you great SEO to get your website back on track, but for someone wanting to retrieve their previously high rankings following a domain change, this can be of little consolation.
So, in the grand tradition of extracting whatever triumph you can out of the current disaster, we have decided to use our website and domain migration as a case study:
"How long does it really take?"
So we will be reporting monthly after the go live on how things are going, and, if client work allows, posting up on this blog as a useful exercise in tracking the time it takes for a website to recover its Google Rankings.
So today before we start the process I needed to work out what constitutes recovery, and attach some figures to that.
I should add the caveat at this point that we have rarely done any real SEO on our own site as the pressure of client work has, like for many companies, meant that we haven't done the work that we might like on our own rankings. Plumbers and taps, cobblers and shoes, as the sayings go. But although we haven't done much SEO we have ended up with a domain ranking reasonably well, thanks to fairly regular blogging (but never enough, as you will identify with yourself), and to our extensions module issued to the Joomla community, some reasonably active social media accounts and also to a good number of links that have been created to our domain by websites we have built. We do have contol over a number of those links, which I confess does give us an advantage over many websites that would not be in such a position.
So to the figures. Let's look at the raw data as of January 9th 2013:
|Google Page Rank||4||Shame, it used to be 5. (Should have done more SEO)|
|Number of Pages||6,580||According to Google|
|Number of Backlinks||1,173,630||According to Google Webmaster Tools|
|or...||426||According to Alexa (Note the delightful variation in SEO metrics reader, this is normal)|
|Alexa Rank||781,586||A blunt measure of visitor traffic, where No 1 is the world's biggest website (Facebook)|
|Number of Unique Visits||3,549||Google Analytics - All of these figures are monthly for December 2012|
|Of which from the UK are:||2,125|
|And next Holland:||659|
|Unique Visitors||2,978||Google Analytics|
|Pages Per Visit||2.25||This will be useful once we look at the new site and determine whether people like it more or not|
|Bounce Rate||69.48%||..as above|
|% new Visits||80.84%||Combines SEO effectiveness and visitor retention|
There are actually a number of other metrics that would be useful to add to this table, but I will pull a few of those out later on.
For those who are curious, our website shows an unusual imbalance. With most websites, the home page is the most powerful and high ranking. This was the case with ours until earlier this year when we issued a plugin for Joomla to help websites be compliant with new EU cookie laws. As it happens the free version of this plugin includes a link back to our website, which goes straight to the page about the EU cookie plugin. As a result, the vast majority of links to our site now go to that page, which has overtaken our home page in rankings and traffic. Now whether this is a good thing for the business overall is debatable, as those visitors are on the whole quite unlikely to be candidates for our services. However, an SEO company rarely says NO to a link, so that is the way things have evolved.
As another aside, I will list our ten main landing pages according to Google Analytics. That is not really what we will be reporting on, but you can see the large effect of the EU cookie plugin page and links. We can take some comfort from seeing that despite the large number of links to the cookie page, we are still gaining slightly more clicks (170 in December) to our home page than to the cookie page. Note also how the Blog post about Google Alerts has shown a lot of times since it accidentally gained a Page 1 ranking for "Google Alerts", but how low the CTR (Click through rate) is to our site for that phrase, compared to our name. This all makes perfect sense if you think about it.
|Rank||Landing Page||Impressions||Clicks||Average Position||CTR|
A bit disappointed to see I'm not getting a look in there!
Here's another interesting data set - the top ten search phrases by volume for which we rank.
|4.||online marketing courses||250||0||19||0.00%|
|5.||web design cornwall||250||5||17||2.00%|
|7.||internet marketing course||200||0||11||0.00%|
|8.||internet marketing courses||200||0||17||0.00%|
|10.||"to select a category: you can either use the box below, or navigate to your desire category and click submit your site link on the top right corner"||170||0||76||0.00%|
But this is better - the top 10 search phrases that bring us visitors - rather different:
The difference in CTR between the searchers for "Channel Computing" (who are looking for us) and "google alerts" (who clearly are not) couldn't be more telling!
|3.||channel computing limited||5||5||1.0||100.00%|
|10.||create google alerts||5||5||12||100.00%|
But back to the story....
I may fill this post in a little more over coming days as it seems to have got late (sorry gotten late for my American visitor!)
However the main update should come about 30 days after the new site goes live. I will be looking to provide similar data to this for that site after 30 days. You will see all these figures drop through the floor initially - that comes with the territory. But as time goes on and we find a few minutes we will:
- Put in place a system of redirects from old content to new
- Restructure and develop the content of the new site
- Re-point links from the old site to the new
I will be working on this from time to time with the amazing Helen Trudgeon and the rest of the Channel crew, so the next update might not be from me.
And as time goes on we will start to see how long it takes us to recover to the position stated above. I will certainly find this interesting and I hope you will too. We're unlikely to work as hard at this as we would on a client project for obvious reasons, however we do have the advantage of control of some of our own links to balance that out. Please stop back in a month or so to have a look and to give us a little moral support.