There’s a chance you might have read this somewhere else, I suppose, but there’s an election going on today. Which is super exciting, of course (though it has seemed to result this morning in two fewer carriages on my train, which means I’m sitting cheek by jowl with a bunch of noisy enthusiastic kids), so I figured it would be remiss of me not to have some sort of look at what can only be described as ‘the events of today’. Don’t worry – this isn’t going to be some political screed – it’s too late for that, I suspect – but what I will say is this: any party that made an election promise to phase out one pound, two pound and fifty pence coins in favour of one pound notes would have my vote, immediately and for all time.
No, what I really wanted to do is take a look at search patterns during the election. When we get a new client, one of the first things that we do is some keyword research. Often a client will come to us and ask that we target certain keywords, or ask us to insert keywords mid campaign. However, it’s important for both parties to remember that actually identifying relevant keywords is only half of the battle. When performing search engine optimisation (or PPC research, for that matter) it’s also important to look at whether people are actually typing the keywords into their search engine of choice and what sort of competition there is, too. This is no different than doing market research for a new product – a lack of giant, polluting metal coins weighing down your pocket and stretching your jeans might *sound* like an awesome idea (because it is), but is it something that other people want, and is anyone else offering it?
All this being the case, what can this sort of keyword research tell us about the 2010 election and the political parties involved?
I decided to look first at search volumes and Google ad costs for keyphrases around the three main parties. With hindsight, I should have included regional parties, the Green Party (my friend is a party local counselor for the Greens in Brighton, and he may not forgive me for this!) and other minor parties. We may come back to this at a later date, if the results are interesting. As this post is going to run very long, I’m going to take a look at the personality driven side of the campaign sometime next week.
Firstly, let’s look at the search trends for the terms ‘conservatives’, ‘labour’ and ‘liberal democrats’ since the beginning of the year:
Pleasingly, Google trends assigned red, yellow and blue to the parties without me even needing to change anything. I was very excited about this.
For those who haven’t used Google trends before, these numbers are normalized so that the maximum search value is 100 – everything else is shown as a percentage of this. So it’s not showing absolute values, but rather relative values. Now, what this means is nothing more than ‘Labour’ is the most searched term of the three – it doesn’t show that the people searching that term actually *like* Labour, just that they want to find out more, good or bad. How does this compare to the last election, in May 2005?
Again, we see that Labour is the most searched term. What is interesting to me, comparing these two graphs, is that there has been a lot more variance in this cycle. I don’t think it’s too much of a leap to say that the televised debates piqued the interest of many people, and caused them to head to their computers and start searching for information.
Now, let’s look at which search terms are currently the most popular, as estimated by Google Adwords:
I’ve colour coded this for ease of viewing. Now, this will give you an idea of some of the issues that the search marketers for the parties are up against; namely, that their parties are known by a number of names. It might look at first blush that Labour is lagging behind, with only two entries here, but in reality what’s happened is that their searches have been concentrated on two specific terms, rather than being split over a number. A good example of Brand Identity, perhaps. It’s also worth remembering that the three parties are not competing against each other for these named searches – their brand keywords are their own. It’s also worth taking the absolute numbers with a large pinch of salt – I’ve included them for the purposes of comparison, not to say exactly how many people are searching for a particular term.
This is starting to run on a bit now and I have clients to whom I must devote myself now, so next time we’ll take a look at the ad costs for some terms, to try to see what the most competitive ones are.
Just to sum up – this is not intended as a totally comprehensive piece of analysis, but more just to scratch the surface and hopefully act as a conversation stater – is there anything you’ve noticed during this general election about the ways that the parties have started using the internet to funnel voters to their website? It’d be especially interesting to hear from members of smaller parties, as I suspect they might be more at liberty to speak. Please, though, let’s not bring any serious politics into it – that’s for another time and another place, perhaps.
