I’ve never taken the contextualised autocomplete seriously before as I normally type without looking and miss the initial suggestions. But this time I noticed something quite useful.
The auto-complete starts as soon as you start typing. So if you stop at “a” the suggestions show you the most popular searches beginning with “a” – in this case for Amazon.co.uk on 3 June 2012, “A song of earth and fire“.
This made me think about a couple of things:
- The Amazon autocomplete suggestions are not well documented and appear to be the “most searched for” - but this might be adjusted by Amazon to lead purchasers into buying another product (see Evan Rosebrook for details on extracting this using the Amazon API and JQuery)
- Could I extract this information and build something useful
After extensive Google research, I am still unsure about the first. But the second became a project for me and my son.
The Amazon Alphabet
We decided to keep a list of all the search recommendations from A to Z and see if they changed over the course of a couple of weeks. Interesting in itself, but more interestingly, not available elsewhere on the web.
Ever the geek, I registered the domain www.amazonalphabet.co.uk and installed a clean WordPress setup so that the platform ran like a standard blog. After installing Yoast SEO, Statpress Visitors and the GooglePlus theme – I was ready.
My son now has the task of once per week pulling the A to Z autocomplete and updating a new page on Amazon Alphabet - when we have a few weeks data, we’ll make charts of each individual letter.
I’m not sure what use overall this will be, but the list itself is an interesting reflection of what people in the UK are searching Amazon for.
Check out Amazon Alphabet – leave a comment for my 11 year old.
ps – if anyone buys anything from these links, my son will get the affiliate commission — his first step into earning via the web