We love digital - Call
03332 207 677 and say hello - Mon - Fri, 9am - 5pm
Call 03332 207 677
There are many ways to optimise an AdWords account, from improving the structure to implementing ad extensions, but one of the most important things is to improve your results from analysing the data. This is the first in a series of posts about data segmentation.
Some people don’t like to get in to the nitty gritty numbers and get scared by spreadsheets, but the methods I’m going to cover make understanding the numbers easy and simplify your optimisation, and for once, my post won’t require you to open a spreadsheet! Unless of course you want to…
It’s all about the segmentation!
AdWords provide a range of ways to compare metrics to help you decide what works and what doesn’t. From looking at this you can then decide what to target more and what not to target, making your account more focussed and more profitable in the long run.
Basic segmentation in AdWords
There are two ways to access the segmentation options, you can either find the Segment button from the Campaigns, Ad Groups and other tabs or you can show the Dimensions tab and choose from additional segments from this list. All segments are applied to whatever is selected on the left hand side of the AdWords interface.
I’ve split up the types of segmentation in to sections to break the analysis down in to manageable chunks; User, Strategy, Time and Visuals.
USER: Geographic location, Demographics, Reach and Frequency, Device
STRATEGY: Destination URLs performance, Placements, Search Terms
TIME: Week, Month, Quarter, Year, Hour of day, Day of the week
VISUAL: Click type, Top v Side, +1 Annotations
And these are just the AdWords ones, there will be more to come for other areas!
Today we’re focussing on the User.
Understanding your target audience, and your converting audience, is a fundamental aspect of managing any advertising campaign. There are several ways you can assess your audiences through AdWords to see which are making you money and which are not as valuable as you expect them to be:
Even if you’re only targeting one country, there is a lot of variety across different areas of countries. If you have a limited budget you may benefit from only targeting the most profitable areas of countries. This example shows how the Cost per Conversion (CPA) varied across towns from £8 to over £46. It also shows which areas have the best conversion rates, the highest volume and the standard click data is also visible in the report too.
I’ve highlighted some of the most notable points in the image, notice how low the CPA for Exeter is, yet how much it costs in Manchester. The conversion rates shown are all quite close here, but the full list ranges from 0.68% to 100%.
Based on this data, I can exclude my campaign from showing in the low performing areas by using the location targeting settings at campaign level. This is most beneficial when you have limited funds and need to maximise your return.
When you advertise on the display network Google can understand more about your audience and help you understand who really converts and who is costing you too much. This example shows the age and gender of the ads audience on youtube.com:
This data is from campaigns which either show on pages with the target keywords or to users who are known to be looking for the service. You can deduce from the results that the product advertised in this account is mainly purchased by men and that most of these men are ages 18-34.
To save money and improve results you could prevent the campaign from showing to users over 35 and females, in addition to the other targeting settings you have in place. To do this, head over to the campaign settings and near the bottom you will find the Demographic bidding section, edit this to exclude your least profitable audiences.
Reach and Frequency
Another good segment for the display network is the Reach and Frequency report – this shows the differences in results between users who see the ad once, twice and all the way up to 8+ times, per day, week or month. It can help you decide what to set the frequency cap of your display ads in each campaign to, in order to ensure users are seeing your ads the optimum number of times.
This example shows that users seeing the ads twice had the highest Click through Rate (CTR) but those who saw it three times had the highest conversion rate:
AdWords allows you to target computers, mobiles with full browsers or tablets with full browsers through the campaign settings. Within the mobile and tablet segment you can specify a narrower target including operating systems, network operators and whether or not to target users on Wi-Fi.
One you have selected your targeting options and run campaigns on these devices you should then start analysing the difference in performance across each platform. In a number of business sectors, mobile and tablet traffic is currently not as competitive as it is on desktop computers, making it cheaper and often more profitable for you, see the details behind this in my post about investing in a mobile website or app.
To see the data in AdWords, pick the tab you want to analyse (campaign, ad groups, keywords etc) then click Segment and choose Device. Here is an example of what sort of data you might see, in this example mobile is the best performing device. Knowing this you may want to review how your site works on tablets, as it is not getting as high a conversion rate as mobiles. You might also want to invest more in mobile as it is giving you a better return for your money.
Improving the data further
If you can’t immediately see the highs and lows that you’re looking for, I recommend downloading the data and opening it in Excel (yes, I know I promised you wouldn’t need it in this post, but it’s sooo useful!). Once you have the data in Excel select it all and apply conditional formatting to make it look as pretty and easy to understand as this:
Look out for information and tips on other segments in future blog posts!
Unlike 08 numbers, 03 numbers cost the same to call as geographic landline numbers (starting 01 and 02), even from a mobile phone. They are also normally included in your inclusive call minutes. Please note we may record some calls.