I’ve mentioned many people’s misplaced obsession with search advertising keywords before, but it’s an issue which keeps coming up. As someone who manages Google AdWords accounts, I’m often asked by clients what keywords are being used, but almost never what the adverts currently look like. This is completely the wrong way around, so let me explain.
(To be fair, one client said: “I know what you mean, Chris, but it’s my MD who wants a report of what keywords are being used, and I really don’t have the time to explain to him why we should be spending time on other things”. Maybe this article will help).
Let’s assume we’re in a world where people read your advert diligently and never click on it unless they’re really interested in what you have to say. Also, let’s assume that clicks on your advert cost the same, whatever search results they appear next to. Remember, that if they don’t click on the advert, you pay nothing.
So where would you like the advert to show: on the search results page for ‘blue widgets’, or the search results page for ‘blue widgets’ and the search results page for ‘red widgets’?
Obviously, it’s the latter. Why wouldn’t you want it to appear twice as often?
What about on the search results page for ‘leicester city’ or ‘eu referendum’? Well, why not? It’ll be seen millions more times. It won’t cost you anything unless people click on the ad, and they won’t click on it unless they’re really interested in what you have to say. So let’s go! Some of your prospects will be making those searches, and your ad might catch their eye.
In fact, in our ideal world, we’d want our advert to appear on every single search result page delivered by Google. Instead of our advert showing a few hundred times a day, we could have it show 3 billion times a day!
So why do we restrict ourselves to just a microscopic set of search results pages, as defined by our ‘keyword list’? Why do we tell Google to only show our ad on about 1 in 10,000,000 searches?
Well, there are two reasons. Firstly, if everyone did this, there’d be over a million adverts on every search results page. So Google discourages us from advertising against searches where we don’t get a high ‘clickthrough rate’ by charging a very high cost per click there. We can still do it, but it doesn’t make a lot of financial sense.
Secondly, there will always be accidental or mistaken clicks. The more pages where your ads appear, the more of these there’ll be. Suppose you’re advertising your blue widgets on the ‘blue widgets’ search results page (very relevant) and the ‘red widgets’ search results page (not so relevant). Let’s say every 100 times the ‘blue widgets’ search results page shows, your advert gets clicked on 5 times correctly and once by accident. And every 100 times the ‘red widgets’ search results page shows, your advert gets clicked on just once correctly but still once by accident.
If a click is £1, then on the ‘blue widgets’ search results page alone, you’re paying £6 for 5 good clicks. On both pages combined, you’re paying £8 for 6 good clicks. Not quite as good a ratio, although potentially still a good RoI.
However, going back to the first reason for not advertising everywhere, as we don’t get a high clickthrough rate on ‘red widgets’, our clicks might cost twice as much. So in practice we’re paying £10 for 6 good clicks. That extra good click from the ‘red widgets’ results has put the price up from £6 to £10.
You can imagine that effect magnified hugely if we were on millions of irrelevant pages. But there is a balance to be had, and many inexperienced advertisers err way too far on the side of caution. In engineering and scientific marketing, there aren’t lots of searches to go round, and the key to success is to keep trying out new ones, while keeping on top of the data.
Don’t think: “Does that Google search results page represent exactly what my company sells?” Do think: “Might my prospects be looking at that page?”
Because if you think the only Google results page which a potential blue widget buyer ever looks at is the results page for blue widgets, you need to expand your horizons.