Step Away From The Numbers

A long time ago, in a galaxy far, far away, or maybe just a matter of years in reality, marketing teams worked tirelessly to target advertising and products at their chosen market – they ran focus groups, conducted psycho-analytical studies and carried out all other manner of tests to try get into the heads of their potential customers and fill in the unknowns about how effective their campaigns were, and also to work out how to better tailor their efforts.

In search marketing we have wealth of data concerning which keywords did this, which ad copy did that and what time of day produced the other. Using this data we can be ever more efficient with given advertising budgets, without having to give two hoots about who is buying the products – we can just follow the numbers.

It's my opinion that it's all to common to create paid search campaigns which are a bit black-and-white, based purely on a data and created using automation to find and combine keywords. Ignoring some of meaning of 'marketing' in such a manner can really hinder the potential of a campaign.

“A good decision is based on knowledge and not on numbers.” - Plato

One further negative of this methodology is that it can lead to PPC being siloed as a channel, often being excluded to some extent from the overall marketing strategy discussions.

Opinions of paid search are often tarnished by the view that it is a fairly one-dimensional channel to which a certain amount of budget is allocated, and it will return at a rate of around X:1 (the return being incremental of course)

I firmly believe that the more a PPC account manager is clued in to a businesses overall goals and strategy, the more successful the campaign will be.

Drawing Parallels

I often talk to clients and other members of the team here at QueryClick about how important identifying underlying user intent is when considering target keywords, and I feel this is a core principle that binds search analysis and old school market analysis

Taking A Fresh Approach

When it comes to creating a new account, it is all to easy to scan the keyword lists that have been spat out by the Google Keyword Tool (other keyword tools are available), look for commonalities within the strings of text and ad group them up accordingly.

This is all fair and well, but couldn't we refine this process somewhat by including some more intelligent thinking? As in more traditional marketing, there is a great deal of insight that we should be collecting before any keyword research has done.

For example, when most sites or businesses are conceived, they should have personas which distil their potential customers into a concise summary, which is then used to guide the overall concepts, design and marketing strategies.

If the paid search keyword research were to extend from here, trying to evaluate the types of keywords which each persona may be suing, we would naturally end up with extremely well targeted campaigns, giving us ad groups and campaigns which keep all the keywords for each persona segmented.

In turn this allows us then to follow the thread through, and develop ad copies to populate each persona specific campaign which can be written to suit the wants and needs of the potential customer laid out in the persona document, taking some of the guess work about what will/won't work in terms of driving click-throughs and ultimately conversion.

The result of this being a campaign which fits the bill in terms of driving the visitors which fit into the original brief for the target customers.

So, I urge all paid search managers to step a way from the numbers every now and then (I know it's hard, I love the data as much as the next man) and take a look at the bigger picture of what you are trying to do. Yes, the data is hugely important for optimizing ROI and it is how we are judged at the end of the day, but there is a real chance to improve your results by remembering the discipline is call search marketing.