Selecting Keywords
Unfortunately this is not a unique post, in fact it is something I have harped on about a lot before - but it seems extremely relevant to me at the moment as I have had two incidents recently where I have seen that people don’t understand the concept of keywords with regard to visitor numbers to a website and more importantly the business purpose of the website.
Let me explain; your website exists for a specific purpose, whether it is to sell bed nights for your hotel, or tours, or houses on a real estate site, or rental properties as in the case of the property development guys I had lunch with yesterday, or in the case of many sites, just as source of information. The biggest mistake is to get trapped in the game of focussing purely on visitor numbers, that is, all search engine and marketing effort you put in focusses just on increasing visitor numbers, without a commensurate focus on real website purpose - say for example selling bed nights. Let me put it another way, if Xebidy reduced the overall number of visitors to your site but doubled the number of people booking a nights stay in your hotel would you say that the search engine and Internet Marketing strategy was a success or a failure?
Hopefully, a success - the return on investment is better. So, one of the most important steps in this regard is the selection of keywords that you will focus on in your marketing efforts to attract customers to your site. This is not rocket science, identify your target market, if you sell tours in Northern Territory to 35+ age group then targeting keywords such as “backpacker bus” and “backpacker tours” is a waste of time and effort, the result is that you will have a high bounce rate on your website. That means, visitors are coming to your site from search results expecting to see some information based on their search query and instead seeing something different so therefore immediately leaving your site. Step 2, identify the search queries that your target market is using, assuming then that you appear high up in the search results page for your selected terms, the higher the search volume the better. If you can find that getting high rank results in search engines for a particularly competitive terms (say Sydney hostel) is difficult, then target more specific terms that might be easier to rank with but nonetheless are more likely to lead to conversion (purchase etc) of visitors on your site. Studies have shown that the more words a user puts into their search query the more likely they are to be a purchaser.
Finally, the keywords you select should be proprietary information, they should be your competitive advantage. If you approach your keyword phrases with a great deal research and then refining them ongoing based on your analytics, which terms lead to more sales, which terms don’t - this is not something you should be telling everyone. We can use lots of tools such as Spyfu to try and work out what competitor keyword terms are, but at the end of the day only you know your sales figures. More importantly, you should be setting up goal conversions and funnels in your analytics packages so that you can see what customers are doing on your site when they come to your site from particular search terms and identifying which ones make you money.
Let’s sum up with a hypothetical example, Bob has a tour company running backpackers around New Zealand. 40% of Bob’s market is Americans, 40% English, 10% Canadian and a smattering of other nationalities thereafter. Bob’s competitors all rate very highly for the terms Backpacker Bus and Backpacker tours - Bob doesn’t. But Bob does do well for some nice generic terms such as New Zealand and Queenstown, and even adventure travel, he has spent a great deal of money targeting these terms. But, he hardly makes any sales online - why not? Because, when someone comes to his site from the search term New Zealand they leave it again. They were not looking for a backpacker tour they were looking for information on New Zealand. Xebidy do the analyse of the search terms with Bob and discover that before his market travel to New Zealand 90% of them do not use the term Backpackers. In the US Backpackers are people that go hiking in the bush and camping. In fact, Bob discovers through his analyse of his market that most of his American market are students and that the main search terms they use are in fact budget travel, student travel and student vacation. Wow and behold, this is a long way off from “backpacker”. Bob reworks his strategy, suddenly his web sales to his America market sky rockets. He does the same for his UK and Canadian market, identifying the keywords that they use in search engines and targeting his site to match. An interesting thing occurs, Bob receives the same amount of traffic as before but from different sources. He receives less customers reaching his site for non-specific terms like Queenstown and more traffic for specific terms like “cheap bus travel New Zealand” and his sales go through the roof. The last thing he is going to do is announce to his competitors, hey I found where all the business is!
Sounds too simple? It is! But it is staggering the number of people I talk to who don’t get it. They want to target broad keywords terms that don’t generate sales, that try to be everything to everyone, and interestingly enough they don’t identify their market. The use of “bus” for companies selling tours, “hotel” for companies selling hostels and Queenstown Restaurants for a company that leases commercial space properties in Queenstown. Go figure!

February 21st, 2008 at 3:17 pm
Naturally every different online marketing channel (SE’s, adwords, forums, directories) has it’s own quirks - but this is taking it too far isn’t it?
I actually think this is an error of basing phrase selection on Analytics and not looking at what people would actually use.
If that sounds like an oxymoron - it’s not, as Analytics is only as good as the historical information that it has.
Also dismissing a keyword because it doesn’t convert well is rash - the first place to focus on is the usability and information architecture of the site.
While it’s a given that you wouldn’t target a phrase that has nothing to do with your product or target market - that doesn’t mean you tighten your keywords/phrases so severely that you lose sight of the diversity of your market.
The first thing you learn in SEO is that you need to think like your customers.
For us:
- Even if our vehicles aren’t strictly ‘campervans’ - thats what our customers search for.
- Even if we don’t like the phrase ‘bus tour’ to describe our service - thats what our customers search for.
Looking at the site you are writing about:
- Nothing, from the photos of the 20 something year olds taking surf lessons to the tour descriptions suggest it is 35+ only
- If describing the tours as “catering for the active budget traveller, backpackers… ” and for “the fit and adventurous who want to get off the beaten track ” doesn’t warrant the targeting of the phrase ‘backpacker’ I don’t know what does.
- As does showing photos of “buses on tour” - and then not targeting the phrase ‘bus’ on the site.
I just can’t understand why you would say that this company doesn’t warrant the term ‘backpacker’ in its SEO strategy? Or ‘bus tours’?
Why would you turn people away who are searching for this?
Yes - the bounce rate ‘might’ be higher (though this isn’t necessarily to do with the keywords - could be the site structure) - however that group is very close to your assumed target that even at 50% of normal conversion it would be worth it (not to mention the word of mouth and other social networking advantage it could have).
At the end of the day you can easily target a number of groups - without compromising your sites message in any way.
It’s not like it has to be an either/or scenario - which is what your example of “if Xebidy reduced the overall number of visitors to your site but doubled the number of people booking…” seems to suggest.
Why in the world would you have to reduce the overall number of visitors to increase bookings?
While compromises do occur in terms of keywords/phrases selection - these should only ever be at the lower end - not on relevant phrases that bring in traffic.
The strategy you mention above actually sounds more like the refinement of individual adword campaigns than SEO.
BTW: Any SEO pro can quickly define the keyword targeted on any given site - there is nothing secret about that.
The only thing that is secret is what someone is actually ‘trying’ to target.
February 21st, 2008 at 3:31 pm
Hi Andrew, my examples were completely hypothetical (don’t read too much into the posts) but you are spot on when you say if the user searches for camper vans or buses then you have to use those words regardless of where you think the product is; and that supports my post example where once you discover that although in NZ and Australia we call them backpackers in the rest of the world they call them hostels and so on. I also don’t think there is much difference between an adwords strategy and an SEO strategy (and you pointed this out recently when you said in a comment that we do lots of A/B testing in adwords, but not in site design and content) they both aim at generating more qualified traffic that ultimately converts to real business. I was at Google in Sydney last week (post coming soon) and we actually talked about how adwords campaigns are most successful when they focus on supplemental search phrases that complement the organic results strategy.
February 21st, 2008 at 3:57 pm
From my perspective - the main difference for adwords is that:
1. You pay for each click-through
2. You pay more for competitive terms
3. You have a daily budget
A very simplistic example could be:
If you had a $10 per day budget, you might have the choice between:
- Targeting “Australian Campervan hire” and getting 4 clicks per day @ $2.50 a click
or
- Targeting “OZ Campervan hire” and getting 20 clicks per day @ $0.50 a click
Naturally there are many other factors at play here including whether you can actually spend your entire budget on the cheaper words and most importantly - which choice brings better overall ROI.
However - for me this drastically different to an SEO strategy, in which you vigorously go after the more competitive keywords - while covering your bases by having the less competitive ones ranking as well (once again just a simplistic example).
Anyway - look forwards to your post re: Sydney.
May 25th, 2008 at 5:26 am
qualified website visitors…
…