How Related is Your Keyword to Conversion in PPC?
As digital marketers, we haven’t been asking this question enough.
There is a swamp of information and products out there explaining how to improve conversions through better keyword bids. A true swamp, but one that leads us to forget that keywords don’t buy things, people buy things.
Take buying a house for example …
Conversions don’t happen online, which is why new conversion points prior to sale have taken prominence in this area. The keyword’s influence on the end purchase is so small that managing your bid to vary performance likely has negligible impact on efficiency.
So, how should you effectively manage your Paid Media activity for products where the keyword does not take centre stage?
How to Look at Low Conversion Volume Keywords in the First Place
Most bid management platforms take a slightly different approach when dealing with low conversion volume keywords.
Take Marin for example.
Marin makes use of the hierarchies within your folder e.g. “ad groups” and “campaign”, to bundle keywords that do not have enough conversions on their own for you to make a granular bid decision.
This methodology can help resolve the lack of bidding data but does not differentiate your individual keywords within each bundle.
Another option is to bring back the point at which you consider something a “conversion”. This proxy conversion in our example above would be the production of the lead, rather than the full house purchase. This allows for the removal of outside factors such as the customer’s financial viability, which as a marketer you cannot control.
In turn, bidding decisions are made based on a wider net, which while may contain some poorer quality leads, contains more total leads. More total leads means more statistically viable data for the bidding algorithm to work off – which means more effective marketing.
Both of these options have pros and cons and should only be considered once all possible methods of manipulating the bid have been considered.
Each bid adjustment option in AdWords can be used to influence conversion efficiency, once you’ve identified a variance in performance within your reporting data.
Let’s take a look at those options in more detail …
How to Use Device Level Adjustments for Better Conversions
For each campaign or ad group, you can access conversion data down to the device your customer uses. You determine the most appropriate level by examining the volume of data within each ad group and the date range you consider relevant.
Sometimes varying the date range can be useful but other times, using a higher level in the hierarchy would be better. It depends on the variance across time, versus variance across the hierarchy in conversion performance.
Wherever there is a variance across your selected group in the device performance (and you have already considered all alternatives to boost the device level performance) you should set a bid adjustment accordingly.
So taking our example from earlier around house buying – if you analyse both your campaign and ad group data, segmented by device, and see there is variation in performance between mobile and the rest (such as mobile CPA is higher) – the easy thing is to set the mobile bid adjustment to account for this variance and bring the performance in line with desktop.
The harder, but more valuable, thing to do is assess the performance of the mobile journey by testing the ad copy and landing page, or by changing the objective to a click-to-call rather than click through (which is more applicable for mobile in this instance). Adjusting these factors first to try to improve performance may narrow the variance and mitigate the lost opportunity.
This helps highlight that the bid for the keyword is not the only driver of performance and we should consider what the other drivers could be, before making any keyword bid changes.
How to Know When the Hour of Day Bid Modifier should be Used
There is normally a variance in efficiency when it comes to the “hour of day” bid modifier. Used for the right campaign, it can definitely improve results.
There’s a pitfall to look out for in that if your campaign spans different time zones and countries, the hour of day you set for the UK will be way off the mark in Dubai. Obviously, a well-structured account should not have campaigns spanning countries or time zones, so this can be avoided.
A general rule of thumb I use is this: If you cannot think of a possible reason for the variance then gather more data to be sure the adjustment is likely to be valid. If the data isn’t there, park the hour of day option until later.
How to Identify the Best Day for Your Ads to Show
When it comes to the Day of Week option, it’s a similar story – a good efficiency improvement but again needs careful consideration when multiple time zones and countries come into play.
“Day of the Month” is not an adjustment option – but can be managed by campaign budgets and automated rules. Before you can say for certain that a particular day in the month affects performance, a lot of due diligence needs to go into ensuring other factors aren’t the cause.
How to Use the Location Bid Adjustment for More Leads
The location modifier allows you to set a bid adjustment at variable levels of granularity and the decision to do so should be based on volume of data and whether you can explain the variance logically. It does however have some great upside if done correctly.
Going back to our scenario from earlier, we would see variance linked to location and the applicable house prices in each area. A lower house price area may manifest in conversion variance across locations. When this is the case, analyse the variance until you can confidently determine it is down to location, then adjust accordingly to benefit from the option.
Onto your next PPC test …
It’s not always the keyword that affects your conversion – it could be one or more factors playing a part. The only way to uncover the truth is through testing.
The golden rule here is that you should only really adjust bid modifiers once you’re confident they will have an impact – in other words, when testing of alternative landing pages and ad copy is complete.
Certain factors impacting your performance cannot be explained logically through these modifiers alone – and this is where knowledge of the products helps, lean on your experts (in-house or client-side) as much as possible.
So the next time you are looking to improve a campaign’s performance (and keyword data is sparse) I hope you test the fundamentals first, and make an informed decision on when to apply bid adjustment options like those listed above.
We’re here if you need any more guidance – just join the conversation in the comments or fire a question at us on Twitter @Latitude_Group.