Three weeks ago, I joined the PPC team at White. Let me be the first to put my hand up and say I never expected to end up in digital marketing. I recently graduated from Oxford University with a Masters in Chemistry. But, having hung up my lab coat and specs, I began my foray into the world of digital. Now, three weeks in, having passed my AdWords qualifications and already been given the opportunity to do some hands-on work with a number of our clients, I’m loving it.
A question I’ve recently been asked is “Why digital marketing? Isn’t that a strange choice for a chemist?”
Well, No.
Whilst the beauty of PPC is that it lends itself as a great career for people from a number of different backgrounds, for me PPC is a science. But instead of turning up to a lab on Monday morning, I’m logging into an AdWords account. From there, however, the thought process is much the same…Here’s my take on the science behind PPC.
Step 1: Understanding your Goals
Before beginning any science experiment it’s important to have thought about what you are trying to achieve. The same is true with PPC. What is the goal of my campaign?
Am I trying to increase brand awareness, drive conversions or traffic to my website? Which Google network does my campaign best lend itself to? Will a text ad suffice or will a visual display have a greater impact?
More importantly, how will this goal effect what I put into and take from the experiment? In the same way the product of a chemical reaction affects the type of spectra you are going to collect, the goal of your PPC campaign will determine the way you will bid and also how you’ll monitor its progress. You will receive a lot of data from AdWords so it’s important to think ahead of time about which statistics define your progress towards your goals.
With the goal decided, next comes…
Step 2: Background Research
We used to use a motto in the lab – “A negative result is still a result” (normally as an antidote to a particularly bad day!) – but, as anyone who has worked any length of time in a lab will tell you, getting it wrong costs money. Whilst inevitable at times, it’s something you’re going to want to minimise in the long run.
The same applies to PPC. Every time someone clicks on your ad you are going to be charged, so you want to ensure that the majority of those clicks are producing positive results for your campaign. How can we ensure this? Well it begins with careful background research.
It helps to consider the what, where and how of your target audience.
What? What is my target audience interested in? What terms are my customers using to describe the product or service in my campaign? What’s the search volume and competition for related keywords?
Where? Where are my target audience? Both geographically and online? Are they spending a lot of their time on search engines or browsing blogs and other websites related to certain topics?
How? How are my audience accessing this content online? What devices are they using?
It’s likely you won’t know the answer to a lot of these questions yet, but stopping to consider the answers you do know can help get your campaign off to a good start.
In the lab you also need to consider the cost and availability of your raw materials. For PPC your raw materials are your keywords. For each keyword we need to consider the relevance (to both your campaign and your audience), popularity and cost, and then weight each factor with the keywords likelihood of delivering a good ROI.
Finally, like combining your raw materials with your catalyst, solvent and conditions, how will I combine my keywords with my bids and ad creatives to create that winning combination?
Step 3: Setting Up the Experiment and Analysing the Results
With your goals decided and your background research completed it’s time to set up your experiments and start collecting data.
It’s helpful to use a systematic approach to setting up campaigns – grouping your products (landing pages on your website) and keywords into categories or ‘ad groups’ and then writing individual ads and ad extensions with those keywords in mind. Along with improving overall tidiness, this also gives you a good shot at securing that all important good quality score straight off the bat.
When monitoring your results, pay careful attention to those areas that are related to the goals you’ve previously decided. As your campaign progresses you’ll also be able to piece more information together about your target audience answering the what, where and how questions in more detail.
Step 4: Optimising – Testing Variables
If I can pass on just one thing gleaned from the world of science it would be this: When optimising your experiments be careful not to change too many variables at once.
You’re going to want to use the data you’ve collected from step three to improve your campaigns performance. This is best achieved through a measured temperament and a series of small changes.
There is no point going all out and adding some more keywords, changing match types, altering your bids, placements, ad text, landing pages and targeting options. With this method, you’ll be unable to determine the cause of any subsequent change to your performance, good or bad, and as a result any future optimising, will be guess work.
