Adwords Search Funnels To Change PPC Landscape

Posted by Steve Baker on March 25th, 2010

Google, Google Adwords, PPC

It has been officially announced at SES New York that Google has introduced Search Funnels – a new development to its Adwords program that could potentially change the face of the PPC landscape for brands and agencies alike. This means it has never been more essential to ensure the right person is managing your Adwords account.

The process of buying online can often be quite long and complex. If you want to buy a toaster, you’ll probably start out with little or no idea of exactly what you want. So you decide to look at toasters – you type ‘toasters’ into Google, and look on a few websites, comparing different toasters.

Eventually, you decide that you want a Morphy Richards 4-slice toaster, so you search for this. You click on a number of paid search results, before deciding that one of the websites is the cheapest, selling the toaster that you want for £24.99.

You decide to think about it, and later on that day, you come back to the website by searching for the website by name, and buy the toaster.

This is a fairly standard sales route – somebody searches for a generic, short-tail keyword, then a more specific, long-tail keyword, then finally a brand.

The problems begin when you come to analyse your data. Google Analytics and Adwords record the sale against the brand name, so it appears that your brand is the most profitable keyword that you have. What it doesn’t show is that without the other keywords, the user would never have searched for your brand name.

It was to resolve this problem that click attribution was invented. Instead of ‘giving’ 100% of the sale to the final search, you can apply part of the sale to each search. This can be expanded, to include e-mails, banner adverts, or any other medium that you can track – these are called touchpoints.

Google are in the process of rolling out Search Funnels, a suite of reports to help users understand the role of their keywords in the purchase process more fully.

At present, the last click claims the sale every time, which clearly benefits brand name terms, but penalises the other stages in the process. This information will soon be freely available to every Adwords user, and if used properly, it should lead to some significant changes in their bidding strategy.

It will show that in most cases, generic keywords are being undervalued, and that brand names are overvalued. It seems inevitable that since generic keywords are the ones most commonly searched for, assigning them extra credits for their part in the sale process will make them more valuable, and hence bids will increase.

Clearly, as a company that advertises on Google, knowing this is very important. If you were the company bidding on toasters in the example, then you’ve probably been basing your bids on the conversion rate of the keyword. Now that you know that these clicks have been more valuable than you’ve believed, increasing your bids on them will make you more sales, and hence, more money.

However, there are problems with this scenario. Consider a very competitive sector, such as gambling. It’s a fair bet that the moment this information becomes available from Google Analytics, everyone that doesn’t already use click-attribution models will adopt them quickly. There are no more new players out there, and every advertiser (or at least most of them) will start increasing their bids. The result in this case is likely to be that Adwords spend goes up, and nobody gets any more customers.

The flip-side would normally be that you cut the bids on the last-click keywords. But if these are your brand names, the chance are that you are paying very little per click already – the Quality Score will see to that.

But what is the alternative in the above scenario? If an advertiser doesn’t increase their bids, they’ll drop down the search results, and get fewer players (at the same cost per click), so they still lose money.

This is just one scenario, where everyone has deep pockets, and the time/will to use click attribution. This isn’t necessarily the case in all sectors, though.

What happens if you have a limited budget, and can’t push up your bids? Perhaps in these cases, you’ll see an improvement in your conversions as you adjust your bids – increasing some, and reducing others. But if other people are increasing their bids as they have sufficient budgets, then they are liable to move above you in the search results.

The impact of this is a little more difficult to predict. You’ll see more conversions by virtue of adjusting your bids relative to each other, but you are likely to drop down the search results. On the other hand, your competitors with deeper pockets will be paying more per click, but getting more conversions, so they could well be winners.

There is another complication, which makes it almost impossible to predict the impact of widespread use of click attribution. To this point, no mention has been made of what percentage of the sale to attribute to each click. In the ‘Toaster’ example, should you split the sale equally between the three keywords? Should you assign the bulk to the first keyword and last keyword, as these ‘put people onto your scent’ and closed the sale respectively?

There are no answers on this at present. Perhaps as a result of widespread use of click attribution, more study will result in certain models becoming the standard.

But the impact of varying these weightings can potentially be huge, in terms of your optimal bids. With every advertiser using different weightings, perhaps the way to ‘win’ from click attribution is to continually test different models, and identify the most profitable. The relationships between the different attribution models that different advertisers adopt has a distinct feel of game theory about it.

What is almost certain is this. Advertisers are going to bid more on generic, large volume keywords, and as a result, Google will make more money. Inevitably, most advertisers will lose money as a result of this, but the most adept will be able to work it to their advantage – never has it been more essential to ensure that the right person is managing your Adwords account.

3 Responses to “Adwords Search Funnels To Change PPC Landscape”

  1. iPodDanny says:

    This is an interesting development from Google, and thanks for the insight provided in the post. Agree that for many highly competitive sectors it could and most likely will lead to an increase in cost for the advertisers as the overall importance of presence throughout the funnel become more important.

    What will be interesting though is how advertisers might start selecting out certain steps in the process, perhaps focusing their efforts on step two or three in the search process. To use your example with the toaster, an online price competitive merchant might choose to bypass the inital exploratory step in the search process for the user, and instead focusing on catching them once they have decided which product/model they are buying. In that scenario the adwords funnel analysis will be helpful in also removing importance of keywords that has been part of the conversion process, but that has not been necessary.

    Will be interesting to see how quick agencies and advertisers will start using this across sectors and industries.

  2. Tom says:

    Excellent post. We manage our own PPC campaigns and it’s a bit of a struggle to keep up with the changes. If anyone has a super-simple guide, besides the ones at google, to using funnels, it would be much appreciated.

  3. Steve Baker says:

    Hi Danny,

    Thanks for the comment.

    It’s quite possible that some people will do that in some sectors, though without detailed analysis, it could be dangerous.

    Some people searching for toasters (in the example) may not search for a long-tail keyword, and will simply return to the website that had the toaster they liked. Without bidding on the original keyword, the sale would be lost.

    This is more likely to be an issue when price comparison isn’t as much of an issue, perhaps where everybody has a slightly different product/service.

    Thanks,

    Steve

Leave a Reply