Everybody agrees that we should test and optimize our campaigns as we go. This feedback loop is core to performance marketing. But, are you testing your audiences and creative in a structured way, with a science-based approach that builds long-term insights and learnings?
How should you plan creative for your campaigns?
When building creative for your campaigns, it's essential to clearly understand the message you want to deliver to your target audiences. In performance marketing, we often work at the end of the funnel, where the most critical messages are around product availability and price. But, those are messages your competitors can also promote. To grow, you'll need to convey more unique messages in your performance marketing to drive consideration and ultimately, purchases.
Your key messages should be founded on the marketing strategy for your campaign. Consider these questions to guide your process:
Who are your customers?
What are the unique attributes of your brand that you can include alongside straightforward performance marketing messages?
Once you've developed your marketing strategy, you can create aligned assets and then set up your campaigns to deliver your message to your target audiences.
How do you apply a science-based approach to your campaigns?
It's essential to recognize that an untested marketing strategy is just a hypothesis about your customers at the end of the day – no matter how well-founded on data and insights – is just a hypothesis about your customers and not a sure thing.
So, we want to treat our campaigns as a scientific experiment designed to test a hypothesis purposefully. Instead of merely running new creative and comparing its performance to past results, we should structure our campaigns to test creatives and audiences against each other.
When planning your campaigns, consider the assets and historical data available to you and first understand what you intend to test at each campaign phase. Focus on quickly answering individual questions with actionable outcomes that will allow you to optimize, rather than trying to answer too much at once. And, think about the realities of the platform you're testing – what are the structural options available to you? How can you account for learning phases for campaigns?
How can you design your campaign structure around this science-based approach?
Typically, you'll have a hypothesis that a new creative message or look and feel might be more appealing to a specific audience than your past creative. You should isolate this particular question – which creative message are you testing, and against which audience?
Ensure that the audience you want to reach is targetable and large enough to test within. Build your campaign structure to target this audience specifically, excluding noise from other potential customers outside your strategic hypothesis. Within that campaign, organize your ad sets or ad groups to test your hypothesis. Don't mix your benchmark or evergreen creative with your new creative within the same ad set. Instead, ensure your ad sets are internally consistent, where all the assets or ads have the same overall creative concept and a consistent message. This strategy allows you to test different messages against each other at the ad set level while allowing variants upon that message within each ad set. Keeping this level of organization will give the platform algorithms more options to work with during testing.
The campaign structures that enable this scientific approach to creative testing also work exceptionally well with the increasing automation of performance marketing channels. By clearly understanding the specific hypotheses we are testing via our campaign structure, we can build consolidated account structures that separate campaigns along key strategic dimensions. The science-based approach guides automation by giving machine learning algorithms clear and straightforward questions to answer.
What is the value of implementing a science-based approach?
By consistently taking this science-based approach to your creative testing and campaign planning, you can ensure that you're always validating your marketing strategy against the ever-shifting marketplace and building genuinely actionable insights that inform your overall strategy.
Are you taking a scientific approach to campaign planning and testing your marketing strategy? If you are interested in learning more about this strategy or need help with your performance marketing program, contact us for a free consultation.