How to set up your campaign tracker

Once your marketing campaign is out in the wild, the next phase begins: measuring its performance, and optimizing to drive ROI. 

It’s not enough to set it and forget it – you need to understand what’s working, what’s not, and why. 

This is where pre-post campaign tracking comes into play.

How to run a pre-post campaign survey

A well-structured pre-post campaign gives you insights into your campaign’s performance, helping you determine what worked, what didn’t, and why.

For your pre-post campaign research to give you anything useful, you should run at least 2 waves – covering the pre-campaign period, and the post-campaign period.

Pro tip 💡
 
“Running a mid-campaign wave can also be super useful! This can give you another metric to tell you how things are going, in addition to your channels’ engagement and conversion metrics.”

Sam Killip

VP Customer Success

After your 2 main waves, it can also be useful to run a follow-up wave a few weeks or months after the campaign. With this you can see if the results you saw during or straight after the campaign have been maintained. This post-post wave is particularly useful for expensive or really important campaigns.

What campaign tracking metrics should you measure?

Your campaign objectives should dictate the metrics you track. Here are some key metrics to consider:

• Brand awareness: Measure unaided and aided recall to see how well your campaign increased brand visibility. These questions can be similar or the same as the brand recall questions you use in your brand tracker. 
• Purchase intent: Assess the likelihood that your audience will purchase your product or service after seeing your campaign.
• Brand perception and association: Track shifts in brand attributes and Net Promoter Score (NPS) to gauge how your campaign influenced customer perceptions. Asking more qualitative questions about how people perceive your brand here can also be a super useful way to figure out if your campaign impacted overall perception.  
• Ad recall: Evaluate how well your target audience remembers your ads and their messages. You can use this insight to inform future campaign messaging, placement etc.
• Engagement: Analyze social media interactions, website visits, and time spent on site to understand how your audience is engaging with your campaign.

Pro tip 💡
 
“The key aspect for questions is keeping the brand metric questions consistent across the two or more waves of the tracker and ensuring they are shown to respondents in the same order both times. Consistency is key!”

Stephanie Rand

Senior Customer Research Manager

How to choose the right audience for a pre-post campaign survey

Generally the ‘ideal audience’ will either reflect the target audience for the campaign, or it will reflect the audience that could be exposed to the campaign. In many cases the ideal audience may revolve around a combination of both these ideas.

Including a control group – people outside your target group or who weren’t exposed to the campaign – can help you figure out if the campaign worked better for your target audience, or potentially better for a completely different audience.

Pro tip 💡
 
“Set quotas to make sure you get a reliable spread of consumers across your target group.”

Nick White

Customer Research Lead

Selecting and screening your audience

The accuracy of your survey results depends on how well your sample represents your target audience. Here’s how to get it right:

How can screening questions improve the accuracy of your research?

Screening questions filter out participants who don’t meet your criteria, ensuring that only relevant and qualified respondents are included in your survey. This increases the reliability of your data and the validity of your conclusions.

Why is setting quotas important in campaign research?

Setting quotas helps you establish that your sample is representative of the population you’re studying. This is especially important when your audience is diverse, and you need to capture perspectives from all key segments.

Pro tip 💡
 
“Consider oversampling smaller but important audience segments to make sure their perspectives are adequately captured in your analysis.”

Sam Killip

VP Customer Success

Campaign tracking audience checklist

There are a bunch of factors that will go into your decision about campaign tracking audiences, but it’s worth asking yourself the following questions:

✅ Is my survey audience relevant to my strategic aims?

✅ Is my survey audience credibly able to see the campaign?

✅ Is my study feasible? I.e can I find enough of these people to take part?

✅ Do you want to go broad, beyond your core target audience, or keep it laser focused?

Once you’ve answered these questions, you’ll be in a much better position to set your audience for the pre and post tracking!

What’s the ideal campaign tracking sample size?

Generally, we’d advise a base size of at least 385 for a single survey of a tracker, and potentially larger depending on who you’re speaking to and if you’re running the research in a larger market such as the United States.

Your specific sample size will depend on a couple of things:

• The feasibility of your audience – this is how likely you are to get a good amount of responses from your target audience. If you’re running a super niche campaign – let’s say, men between 45 and 47 who go fishing on Wednesdays – you might struggle to get 385 of these people to respond. If your campaign’s targeted more broadly, you can get away with a bigger sample size. 
• The geography of your campaign – similar to the men who fish, if your campaign’s target geo was small and specific, you might struggle to get a sizeable response rate so a smaller sample size will be better. For a nationwide campaign, or one covering a large area, 385+ should be perfectly feasible.

Pro tip 💡
 
“It’s worth bearing in mind the standard statistical rules around margin of error when comparing two self-reported awareness scores, for example, on a base size of 100 the margin of error is 10%, i.e. any reported score could be anywhere between 10% points above or below the reported score.”

“However, for a sample of 500,  the margin of error shrinks to just 4%. This means the results you have are more likely to be accurate.”

Sam Killip

VP Customer Success