Better PPC with Google Attribution Modelling
Goede stof, om de werking van google Attribution 360 te doorgronden.
5 sept. 2016 by William Cheng
In search engine marketing Last Click attribution has always been the default and this is especially true in Google AdWords and Bing Ads pay-per-click advertising. However, this way of viewing your data does not always give you the full picture and can potentially be misguiding your online strategy and how you optimise your campaigns.
This can happen in a number of ways:
Scenario 1
Have you ever looked at the Keyword Report in your AdWords account and seen 20% of your keywords driving the conversions and the other 80% just spending money but without ever converting?
Naturally, you might think that 80% of your keywords are only wasting your marketing budget and so you turn them off but once you’ve done this, you see that the 20% of keywords that were performing well are suddenly not doing so good.
Scenario 2
Or, have you ever compared your Device Performance and saw that Mobile traffic is driving large volumes of traffic but conversion rates are dramatically lower than desktops & laptop computers? Armed with this information you make the decision to opt out of Mobile but suddenly your desktop & laptop conversion rates also fall.
Scenario 3
A lot of people might argue that advertising on the Display Networks is only good for branding and not very strong for direct response marketing. However, when you turn off the Display campaigns, your Search figures also drop.
Let’s try to explain what is happening and what you can do about it, keep reading.
GOOGLE ATTRIBUTION MODELLING
The customer journey has multiple touchpoints along the way, however, with Last Click attribution, only the final touchpoint gets the credit for the conversion or sale.
The image below shows the search journey where the term “nexus 9 tablet” appears to be driving all the conversions but actually “great tech gifts” started that journey and “highly rated tablets” kept it going.
When only using Last Click attribution, end of funnel keywords often receive inflated credit for conversions & sales resulting in over allocation of budget to these keywords and higher CPA whilst limiting exposure to your potential market.
Advertisers might miss out on an opportunity to leverage a much larger market:
Bidding on keywords that initiate or assist conversions should result in a more efficient budget allocation, capture a larger share of the market and achieve optimal CPA.
Google’s new Attribution Modelling enables advertiser to reallocate conversion credit using alternate models. Below shows how the credit would be split across a Position-Based model.
Here is a full list of Attribution Models available:
• Last Click
• First Click
• Linear
• Position-Based
• Time Decay
• Data-Driven
Google believes that the Data-Driven Attribution Model is the best one to choose because it uses sophisticated algorithms to evaluate all the different paths in your account (both converting and non-converting) to determine which touchpoints are the most influential.
However, this only works if you have enough data for the algorithms to analyse, say at least 20,000 clicks and 800 conversions in the last 30 days.
For everyone else, here are some tips and considerations for choosing which Attribution Model best fits your business:
• Always use Data-Driven attribution if you have enough data
• Avoid using Last Click and First Click models as they will almost never be a good fit
• Consider whether you have a an aggressive or conservative growth strategy as the model you use will surely impact your optimisation
• What is your market position? Are you the market leader or new to the market?
• Do you have many competitors or only a few?
• Is your path to conversion short or does it have long consideration times?
• Is your primary goal to grow your customer base or maintain the relationship with your current customers?
Contact Circus PPC today to discuss which Attribution Model would be best suited to your business and how to implement it in your AdWords account.
Written by William Cheng PPC Director at Circus PPC Agency