Using Google Insight for tourism and travel research
I recently wrote a post on using Google Trends for tourism and travel insight and this post expands on some of those themes by talking about the Google Insight product.
Before I do so, it might be as well to remind ourselves of the difference between Google Trends and Google Insight. As I noted in the previous post, “…Google Trends shows data relating to traffic to websites while Google Insight shows data related to search terms.”
So, in a nutshell, Google Insight offers a great way of understanding how people search for particular terms and, more importantly, the contexts in which they do it. For a tourism destination, for example, this means that it is possible to judge where your destination lies in comparison to competitor destinations and whether there are opportunities to broaden your market offering. For a specific tourism or travel business, you can capitalise on the fact that brand names are increasing dominating searches in order to see where you stack up against competitor businesses.
It’s probably best to illustrate this with a concrete example and for this I’m going to look at some tourism businesses in Aviemore, a destination that offers year round outdoor activities close to where I live. I’m going to concentrate on two businesses – the Aviemore Highland Resort and the Hilton Aviemore. I have selected these two simply because they are both large hotels, they both cater for a similar clientèle and they both undertake marketing expenditure.
Inputting the brand search terms ‘aviemore highland resort’ and ‘hilton aviemore’ brings up results that look like this (or click on the image below for a larger version). I’ve applied filters to the results so that I receive data based on the relative popularity of the two search terms from people within the UK in the period Jan 07 through to October 08.
You can see from this graph that the two hotels pretty much shadowed each other up until about July this year when the Aviemore Highland Resort started to drift away downwards from the Aviemore Hilton. Now, there have been periods of divergence before but this recent period strikes me as being longer lasting and deeper than previous splits so, if I were Aviemore Highland Resort, I would now have concrete proof that for some reason, I was no longer making as big an impact when compared to my close rivals. As such, I would either know why (eg marketing budgets might have been changed) or I would be starting to ask serious questions to find out why.
Move from assumptions to proof
But now let’s introduce another search term into the mix to get an idea of whether it’s more a case that the Hilton is performing exceptionally rather than the Highland Resort performing poorly.
In this example, I’ve introduced the term ‘Aviemore Hotels’ as my benchmark term. Whereas the previous terms are brand terms – and likely used for navigational search by people who are already aware the establishments exist, ‘Aviemore Hotels’ is a more open search term that requires no knowledge of existing brands in the area. Therefore it is more of a general benchmark indicator of the broader level of interest in hotels in the area. The result is shown in the graph below (click for a larger version or visit Google Insight here).
One thing that you should notice quickly is that the Hilton seems to trend more closely with the ‘Aviemore Hotels’ line than the Aviemore Highland resort does. Indeed, the raw data enables us to determine that there is a stronger statistically provable correlation between ‘Aviemore Hilton’ and ‘Aviemore Hotels’ than between ‘Aviemore Highland Resort’ and ‘Aviemore Hotels’. In other words, the Hilton is performing in line with the market and the Highland resort less so.
Incidentally, even if we take the figures for Aviemore Highland Resort in isolation, using the raw data (available if you have a Google Account), we can see that the term ‘aviemore highland resort’ is now performing outside of control limits (defined as standard deviation x 3 – see more here about control limits) as shown in the graph below.
As virtually all web sites have cycles, we should expect to see some changes throughout the year but this suggests that the current change lies outside of what might be expected within these cycles:
So what does this mean?
For the Aviemore Highland Resort it means something may be wrong, beyond the level of a mere seasonal wobble.
My first actions would be to look at spend, bookings and occupancy data to see if there has been a corresponding drop in revenues. (Afterall we are just talking about search activity here!)
I would look in depth at web traffic and conversions to identify which visitor segments and traffic sources I have lost search activity and potential business from. I would also look closely at marketing activity and assess whether a drop in advertising spend has lead to this drop in search volume – and whether there is a cost effective way of rectifying that. Afterall, it is common for people to respond to TV and other forms of offline activity by going online and searching on the brand name. Is this what is occurring here – and does it even matter to the bottom line? I’d want to know.
And if I were the Hilton Aviemore? Well, I be heading off to Google Trends and comparing our overall website traffic for clues. I’d be looking at my revenue and web analytics data to see if I was benefiting from this displaced search activity – and whether I was converting it into revenue. And I would bullishly be looking at what I was doing right and be tempted to invest in doing more of the same.
So, Google Insight – used wisely – has the power to act as warning device for your business. It’s free (and this article has only really touched on a small number of its features), so can you afford to ignore it?
Filed by Stephen (03/12/08)
This entry was posted on Wednesday, December 3rd, 2008 at 2:06 pm and is filed under Data, Internet usage statistics, Marketing strategy, Research tools, Tourism market research, Web analytics and web measurement. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.