Why ROAS is a red herring

For any business investing in marketing, calculating your CPA (cost per acquisition) has become a key metric. Put simply, if it costs you more to acquire a customer than you make in profit, you’ll eventually run out of cash.

Ok, now let’s say you’re broadly profitable but you want to see which channels deliver the best returns so you can invest more in them and grow as a business – the simple way to do this is by analysing ROAS (return on ad spend).

So, what is ROAS?

ROAS is a metric that measures the effectiveness of advertising by calculating the revenue generated compared to the cost of running that campaign. It is typically expressed as a ratio, with higher ratios indicating more effective advertising.

What is considered a good ROAS depends on the industry and business objectives. However, in general, a ROAS of 4:1 or higher is considered good, meaning that for every pound spent on advertising, four pounds or more in revenue are generated. 

What’s the issue with ROAS?

There are a quite a few actually. But the main issue with ROAS is that it wants to give all the credit to the last channel in a sale.

Imagine giving everyone in a nightclub queue a flyer for a discount. They were clearly going to come in anyway, but ROAS would suggest that the flyer was the promotional tactic that converted customers, which is just daft and paints a false narrative.

In this fictional example, the nightclub might invest more money in said flyers and the discount, but would the business grow off the back of it? If we’re being kind, we’d say it’s unlikely.

Now, we’re not writing off ROAS as a metric altogether, but it’s clear that it needs to be viewed in context of other marketing and advertising efforts.

Measuring a channel’s incremental lift

So, is there a better way of measuring true success? Well, perhaps take a look at Incrementality. In marketing terms, it refers to the degree to which a marketing campaign or tactic is responsible for driving a desired outcome, such as a purchase or conversion, above and beyond what would have happened naturally or through other marketing efforts.

In other words, it seeks to measure the "incremental lift" in results that can be attributed to a specific marketing effort.

Incrementality analysis can help marketers to optimise their marketing investments and allocate their budgets to the most effective channels or campaigns.

The problem is that incrementality is incredibly hard to measure. To measure incrementality, marketers can conduct randomised controlled experiments - i.e. showing ads to one group while withholding ads from the other - and comparing the outcomes.

This is the easy comparison of users who were and would have been exposed to the ads. But some would say channel-silo incrementality is not true incrementality. They’re right, but it’s a good starting point for coaxing marketers away from their last-click attribution modelling mindset and giving due credit to those channels operating further up the funnel.

What’s the answer?

ROAS has shown itself to be a popular metric because it simplifies - or rather, over-simplifies - campaign measurement.

But if the digital advertising ecosystem is going to mature and evolve beyond its origins as a set of short-term direct response channels into the full-funnel brand-building marketing ecosystem it wants to become, we need to move away from focusing on hard metrics.

When it comes to metrics, there’s no perfect science. CPA as a percentage of average order value (AOV) has gained a bit of traction in recent months, which is essentially Amazon’s advertising cost of sales (ACOS) metric, used to measure Amazon pay-per-click (PPC) advertising campaigns. But it’s basically inverse ROAS.

Meanwhile, customer lifetime value (CLV) remains one of the key stats to track as part of a customer experience programme. CLV is a measurement of how valuable a customer is to your company, not just on a purchase-by-purchase basis but across the whole relationship. But you could still end up giving credit to the wrong channels.

Is retargeting effective? Yes. But would it work without the initial prospecting higher up the funnel? No.

You get the idea.

So, what’s the best attribution model? Well, there’s no 100% right answer but here are five things to help you build better, more profitable campaigns.

• Beware of simple crude metrics like ROAS
• Consider AOV and LTVs
• Look beyond last-click attribution
• Prioritise long-term over short-term metrics
• Brand builds performance

In other words, let’s get brand and performance marketing working in tandem online, rather than simply competing for the same budget.

Contact us today to see how we create and distribute inspiring, on-brand content that performs across your marketing funnel online.