Firstly, you need to understand the impact of spending budget on paid search on other channels. For example when you spend on paid search do you see additional revenue coming through other channels like organic search?
Secondly, you need to be able to quantify the impact of other channels on paid search that might result in additional budget requirements. For example, when you run TV ads you will see an uplift in branded searches that will require additional budget to ensure that you’re protecting your branded terms.
Finally you must be able to model the future impact of adding or decreasing your paid search budget to see how it affects cost per acquisition.
For example if you allocate an additional £500k to paid search you might see a 15% increase in CPA to spend that budget, making the channel less profitable compared to others which may better justify a budget increase.
Quantify the impact of paid search on your other channels
As well as driving direct sales, paid search can also act as the first touchpoint in a customer purchasing journey.
This is where a user first interacts with a paid search advert first then returns to the website to complete a conversion via a different channel such as organic search or programmatic display.
To determine the impact of paid search on other channels we would suggest either using Google Analytics or Google Attribution with a position-based or data-driven attribution model if you have multi-click journeys.
Within Google Analytics attribution you should then be able to view the number of assisted conversions, and assisted conversion value, of paid search on other channels. That means you can take into consideration the value of paid search touch points across varied conversion journeys when allocating budgets.
Quantify the impact of spending budget on other channels on paid search
As well as looking how spending on paid search will impact on your other channels, it’s important to consider how spending on other channels may mean that your paid search performs better and therefore justifies further investment.
A good example of this would be looking at the impact of your TV activity on your paid search performance. A study from BKV showed that 80% of branded searchers for direct response TV were driven by television.
If you’re running TV activity alongside your paid search activity then you should factor in a significant increase in brand searchers compared to average and allocate additional PPC budget. This ensures you can maintain brand presence and protect your brand from competitors.
TV ads can also impact how people interact with your non-branded traffic.The BKV analysis shows an uplift in non-branded visits whilst there was a TV flight running.
This is likely to be caused by greater brand recognition after your ads have run, making your ad on a non-brand search more appealing, and so leading to an increase in CTR and traffic.
When allowing for the impact of other top of funnel channels such as TV, programmatic and out of home, in your paid search budget you should review your historical data.
Review the impact of other channels on paid search on both branded and non-branded traffic. If for example you see an 80% increase in branded traffic when your TV ads are running, allow for an 80% increase in non-branded spend for the length of your TV flight and some weeks after.
Forecast paid search performance
One of the keys to working out what percentage of your media spend you should allocate to paid search is being able to accurately project into the future the impact of making certain changes within your paid search account.
As your cost per acquisition is likely to go up with increased budget then you will want to work out how much it will go up by, and, if at the increased CPA, whether your budget might be better spent on other channels such as paid social or programmatic.
There are several different ways that you can do this. The easiest is to use Google’s new performance planner that was launched on the 14th May 2019 that allows you to model out future performance of your campaigns.
Alternatively you can use regression modelling to model the CPA at different spend levels. Search Engine Land offers step-by-step instructions on how you can model CPA vs spend using regression modelling.
You can then use this data to determine the optimal amount of budget to allocate to PPC before it would make sense to allocate more budget to another channel.
You should be able to competently determine the percentage of your media budget that should be allocated towards your paid search activity, provided that you understand the cross-channel impact of increased and decreased activity in each channel, and are able to create a model to predict impact of budget changes on paid search CPA.