Assisted Conversions: The battle for bragging rights
Assisted conversions, also known as attribution, is something we need to consider more and more to justify the value of paid advertising. In this post, I show you a way of reporting on assisted conversions in a PPC campaign to do just that.
Assisted conversions in an online world
With the growth of online channels and the boom in technology driving e-commerce, we are spoilt for choice and it is easier than ever to shop online. Traditionally, when we shop on the high street, we do not buy from the first shop we go in. We look around, we try to find the best deal for ourselves, or we feel a particular affinity with one shop or shop assistant over the other. Most of us do some digging around, if we don’t find anything better then often we end up back at the original shop. Unsurprisingly, we see trends that reflect this behaviour in customers online behaviour too.
Working with a number of e-commerce brands, it’s important for me to understand the online buying cycle and measure how paid advertising plays a significant part in the process of assisting a sale or conversion such as email sign up.
So, what is an assisted conversion?
In simple terms, assisted conversions occur when a user interacts with a paid advertising campaign but does not complete a purchase or sign up in that initial instant. The user later returns to the website to purchase a product via another source and/or entry point. Typically direct to the brands URL or from an organic brand search. The length of user research mostly depends on the product or service they’re looking to purchase.
For example: If a customer is looking to buy a bed or a car, the buying cycle could be months from the initial research phase; compared to days or even hours for a pair of shoes.
Here is an example situation:
- Google search > click PPC text advert > visit website > leave without purchase
- Look at the brands Facebook page > read comments & reviews > leave to mull things over
- One week later > User returns to the original website found through the PPC advert, but via an organic brand search > User competes purchase
In the situation above, PPC should get credit for kicking off the buying cycle of the 3 interactions, but often it doesn’t. In Google Analytics e-commerce tracking, organic would be attributed the conversion and win the sale.
Now that we know what assisted conversions are, how do we report on them to represent the true value of paid advertising?
Reporting on assisted conversions gives us a clearer picture of the value of particular paid marketing campaigns. My approach may differ from others, but the result should be the same, here’s an example of the process I take to find how valuable a particular PPC keyword has been in assisting a sale.
Say I have a phrase match keyword that spent £213. It received 2 conversions worth £289 revenue, giving me a 135% ROAS. This is below my 300% KPI target. Before I do anything drastic like pausing the keyword, or reducing / increasing bids, I look at attribution reporting in Google Analytics to see if the keyword has assisted any sales.
Under the Conversions section, in multi-channel funnels, there’s a tab called ‘Top Conversion Paths’.
By choosing conversion segments, you can select ‘First interaction is Paid Advertising’. You can now see the true value of the assisted conversions.
This information clearly shows us that paid search has attributed to the conversions and revenue we assumed direct and organic search were responsible for.
Now, I want to segment this data further down to keyword level. You can do this by changing the type from all to Adwords.
Under the primary dimension, we can choose campaign, ad group or keyword path. This breaks down the data further, giving us insights into conversion and revenue at keyword level.
So, regarding my earlier example, I found the keyword which originally had a ROAS of 135% actually has a ROAS of 400% and is profitable and valuable to my Adwords campaign.
If you would like to find out more on PPC advertising and search email email@example.com or call us on 0161 956 8963