Why Brands Should Move Away From ROAS.

Reporting, Ecommerce Strategy

Get new articles directly in your inbox! Click here.

Get new articles directly in your inbox! Click here.

ROAS (return on ad spend) has often been the industry standard for measuring campaign success and the guiding light when it comes to optimising spend. However, a reliance on ROAS (especially last click) causes short term thinking and poor decision making. In any other circumstance than to test, it could be your red tape when it comes to revenue growth.


Understanding ROAS.


Return on ad spend (ROAS) is a calculated ratio which determines how much revenue you are generating from each pound, dollar or euro your brand is spending on advertising. In theory, it provides a clear picture of whether a campaign is successful and worth the money you put into it.


The formula for calculating ROAS is pretty straightforward:


Revenue / Spend


Now, let's take a look at last click ROAS in particular. Last click ROAS means that 100% of the revenue generated by a specific channel is attributed to the last click of the purchase journey. This can be a tricky attribution model when it comes to reporting as it doesn't acknowledge any previous engagement along the purchase journey. For instance, a consumer is served several ads on Instagram and then chooses to Google the brand to make a purchase. They click on a Google ad and make their purchase. With a last click model, 100% of the revenue will be attributed to the Google Ad campaign even though the Instagram ads were the main driver to purchase.


What decision would you make if you had to cut budget the next month? Based on this reporting, you would cut Paid Social budget as no/very little revenue is being attributed against it. But then what would push your consumers to purchase? Nothing. Therefore your revenue growth would suffer.


This example is just one of the ways that ROAS can trick marketers into showing what is working and what isn't.

 

ROAS Has Become An Inaccurate Representation.


So we've already looked at one example where ROAS can be inaccurate. But let's explore some other scenarios we often see which could be limiting revenue growth.


Using ROAS to determine budget optimisation


If your campaigns are organised by location, you will be greeted with a range of ROAS for each country or region. You can use this ROAS, after a period of testing, to see whether it is worth advertising in that specific country. However, as always, this cannot be used as gospel.


Example:


A brand selling luxury luggage is running campaigns in the UK, US, Germany, France and Sweden. Each location has its own budget with the majority being split between the UK and the US.

ROAS for each country:

UK = 3.5:1

US = 4:1

Germany = 3:1

France = 4.6:1

Sweden = 8:1


Looking at these figures in isolation would suggest increasing budget for Sweden as the ROAS is the highest and most yielding. What we can't see is that Sweden only had a budget of £1000 across the month and it took just 2 purchases to achieve at 8:1 ROAS.


The reason Sweden only had a budget of £1000 is because the brand was aware, after extensive market research, that there is not a huge market for their product in Sweden You could increase budget in this region, only to hit a plateau or even negative returns.


ROAS vs Revenue


Using the same example, let's dig a little deeper.


We already know that Sweden's 8:1 ROAS has been generated from £1000 worth of budget.


8:£1000 = £8000. Minus the investment equals a gross revenue of £7000.


US yielded a 4:1 ROAS with a budget of £9000.


4:£9000 = £36000. Minus the investment equals a gross revenue of £27000.


Some marketers will look at ROAS and think that the US is underperforming but it actually generated £20k more revenue that the Sweden campaign.


It is inevitable that as you increase marketing spend, your ROAS will reduce. So a undying focus on ROAS alone will stifle growth. Marketing directors should be looking at gross revenue and incremental profit increases to determine campaign success and leave analysing ROAS to their marketing team or agency. They will be much more attuned to the nuances and contexts around these smaller, less significant metrics.

 


ROAS (return on ad spend) has often been the industry standard for measuring campaign success and the guiding light when it comes to optimising spend. However, a reliance on ROAS (especially last click) causes short term thinking and poor decision making. In any other circumstance than to test, it could be your red tape when it comes to revenue growth.


Understanding ROAS.


Return on ad spend (ROAS) is a calculated ratio which determines how much revenue you are generating from each pound, dollar or euro your brand is spending on advertising. In theory, it provides a clear picture of whether a campaign is successful and worth the money you put into it.


The formula for calculating ROAS is pretty straightforward:


Revenue / Spend


Now, let's take a look at last click ROAS in particular. Last click ROAS means that 100% of the revenue generated by a specific channel is attributed to the last click of the purchase journey. This can be a tricky attribution model when it comes to reporting as it doesn't acknowledge any previous engagement along the purchase journey. For instance, a consumer is served several ads on Instagram and then chooses to Google the brand to make a purchase. They click on a Google ad and make their purchase. With a last click model, 100% of the revenue will be attributed to the Google Ad campaign even though the Instagram ads were the main driver to purchase.


What decision would you make if you had to cut budget the next month? Based on this reporting, you would cut Paid Social budget as no/very little revenue is being attributed against it. But then what would push your consumers to purchase? Nothing. Therefore your revenue growth would suffer.


This example is just one of the ways that ROAS can trick marketers into showing what is working and what isn't.

 

ROAS Has Become An Inaccurate Representation.


So we've already looked at one example where ROAS can be inaccurate. But let's explore some other scenarios we often see which could be limiting revenue growth.


Using ROAS to determine budget optimisation


If your campaigns are organised by location, you will be greeted with a range of ROAS for each country or region. You can use this ROAS, after a period of testing, to see whether it is worth advertising in that specific country. However, as always, this cannot be used as gospel.


Example:


A brand selling luxury luggage is running campaigns in the UK, US, Germany, France and Sweden. Each location has its own budget with the majority being split between the UK and the US.

ROAS for each country:

UK = 3.5:1

US = 4:1

Germany = 3:1

France = 4.6:1

Sweden = 8:1


Looking at these figures in isolation would suggest increasing budget for Sweden as the ROAS is the highest and most yielding. What we can't see is that Sweden only had a budget of £1000 across the month and it took just 2 purchases to achieve at 8:1 ROAS.


The reason Sweden only had a budget of £1000 is because the brand was aware, after extensive market research, that there is not a huge market for their product in Sweden You could increase budget in this region, only to hit a plateau or even negative returns.


ROAS vs Revenue


Using the same example, let's dig a little deeper.


We already know that Sweden's 8:1 ROAS has been generated from £1000 worth of budget.


8:£1000 = £8000. Minus the investment equals a gross revenue of £7000.


US yielded a 4:1 ROAS with a budget of £9000.


4:£9000 = £36000. Minus the investment equals a gross revenue of £27000.


Some marketers will look at ROAS and think that the US is underperforming but it actually generated £20k more revenue that the Sweden campaign.


It is inevitable that as you increase marketing spend, your ROAS will reduce. So a undying focus on ROAS alone will stifle growth. Marketing directors should be looking at gross revenue and incremental profit increases to determine campaign success and leave analysing ROAS to their marketing team or agency. They will be much more attuned to the nuances and contexts around these smaller, less significant metrics.

 


Struggling to acquire new customers at scale? Learn more about how we've helped brands just like yours. Click here.

What Brands Should Be Measuring.


Instead of solely relying on ROAS, there are some more accurate alternatives.


Assisted Conversions


Google Analytics and Google Ads both offer an "Assisted Conversions" report. As we've already touched upon, users will more than likely have several interactions with your brand before making a purchase; so its unfair to allocate all success to the last click. An Assisted Conversions report is a better source of truth across your channels.


Because of the complex path to purchase and The Messy Middle we recommend setting a 90 day "lookback window". This allows us to see which channels have positively contributed to conversions within a 90 day period. If you are running top, middle and bottom of funnel activity, this type of reporting may even show you the most valuable channels throughout the entire purchase journey.


Revenue to Target


Probably the most simple way of assessing success is looking at revenue to target. For CEOs, CMOs and Ecommerce Directors, we recommend focussing your attention here. If your marketing team or agency are consistently achieving or surpassing the revenue target you are setting, then the campaigns are working and that's about all you need to know. If they are consistently underdelivering then it may be time to investigate and dig a little deeper; or your targets are too ambitious for the level of activity you are running.


Econometric Modelling


Econometric modelling (also known as Marketing Mix Modelling) is large scale data modelling and analysis and is primarily used to measure total marketing effectiveness. It is more well known and utilised in TV advertising for measuring the impact of TV campaigns against brand recall and revenue but this type of reporting is slowly moving into the world of digital.


Econometrics is more data driven than traditional marketing and relies on statistical analysis techniques which are then applied to marketing or sales data to estimate the impact of various marketing activities.


A more in depth explanation of Econometrics within marketing can be found here. 


It is not something which can often be set up by an internal team so approaching an agency specialising in Econometrics will be your best port of call.

 

Money Metrics


In terms of day to day metrics you should switch your focus too, try these.


CAC


Customer acquisition cost - Designed to measure and maintain the profitability of your acquisition efforts. Your aim should be to decrease this.


LTV


Lifetime value - How much value each customer will bring over their consumer lifetime. Your aim should be to increase this.


AOV


Average order value - The average size of each order. Your aim should be to increase this.


Gross Profit


The money you've made once you've deducted the cost of goods sold.

 

Looking to The Future.


If digital marketing teams and it's leaders wish to evolve beyond these short term reports and responses, they will need to change the way they work. If all your marketing campaigns are set up and optimising to ROAS as the primary KPI, you will be limiting growth to the disadvantage of the brand.


Marketing should be looked at holistically from the top. That means pulling a greater focus on incremental revenue growth, looking at growth metrics, across both brand and performance and ultimately lasering in on revenue. After all, it is a marketers job to make money not to achieve the highest ROAS.

Looking for

Support?

If you or your brand are looking for support with a particular service or have a question about what you've just read, get in touch and we'll be happy to help!

Looking for

Support?

If you or your brand are looking for support with a particular service or have a question about what you've just read, get in touch and we'll be happy to help!