17 November 2025 Deeply Digital

Driving data-driven decision making with marketing experimentation

Decision making is hard, but inaction is ultimately still taking a decision.

We’re all guilty of sitting on something longer than we should - sometimes it’s fear, sometimes it’s procrastination, sometimes it’s the hope that the decision will simply disappear or someone will make it for you. 

Unfortunately, marketing is full of decisions. They’re not all big decisions, it might be something as simple as deciding when a social update should be posted or signing off email copy - if it’s your decision, then it’s also you that gets held accountable for the consequences, good or bad. 

So how, as marketers, can we avoid the inevitable decision paralysis that we all come across from time to time?

Data-driven decision making

It’s a bit of a chicken and egg situation - to get data, you have to do something, you have to make a decision, but to make decisions, you need the data. 

Firstly, you probably already have some data, so that’s a starting point, and if you don’t have data, then action really is the only way forward. The key here is to be prepared to pause, stop, pivot or “double-down” on the marketing activity you’re running.

As an example, if you choose to run a LinkedIn ad campaign for a new audience, make sure you’re gathering the correct data from the start. What that data is will be aligned with your marketing strategy and goals, be it website sessions, lead generation or post engagement on LinkedIn.

If you can start by tracking the right data, then decisions become much easier to make. Taking this example a step further, let’s say your goal is website sessions / clicks to landing  page - this is a basic metric on LinkedIn that doesn’t necessarily need additional tracking. 

From day one, you should be monitoring this metric closely, along with the click-through rate (CTR). If after a couple of days, the performance isn’t aligning with the overall goal of the campaign (taking into account the forecast and budget over the campaign lifetime), then you might need to pause the campaign.

Data-driven experimentation

At this point, using the data you have to hand, you can take the next decision: stop completely and try something new (possibly even a different channel) or pivot and change something (messaging, imagery, bid strategy etc - just try one, not all at the same time) to see if that gets the campaign back on track. 

In this example, not pausing or pivoting based on the performance you’re seeing is as good as doing nothing, and you can’t expect the results to change during the campaign, using your budget on an underperforming campaign. 

If instead, the LinkedIn ad campaign in the example took off by storm in the first few days, you might choose to apply more budget or if the campaign allows for it, widen the net to try and boost the results. 

This is a simple example of how data can help you make decisions in marketing, in this case running a LinkedIn ad campaign, but the same logic applies to just about any marketing channel or activity. 

Experiment with your marketing

When launching a new campaign or message, experiment by trying different content, different channels, different promotional tactics. The reason this can help with decision paralysis is because it doesn’t rely on one big decision that you feel can make or break a campaign, instead, it provides lots of different tactics, hopefully some of which will stick. 

When experimenting with different ideas, it’s also good to hypothesise. These can be small hypotheses that help you make the decision and provide something to test against. 

An example of this might be a hypothesis based on previous marketing data or market research: By emailing our audience at 10:30 on Tuesdays, our click rate is likely to be higher. This is backed up by data we have collected from sending emails at varying times over the past 6 months. 

With this data, you can be more confident about the decision to send an email at 10:30 on Tuesday. And if it doesn’t work? Note that the hypothesis has been disproven and as a result, for the next few emails, you’re going to test different times and days to build up new data. 

As marketers, we have more toolsets and channels to experiment with than ever before and by leveraging AI, the speed at which we can generate new content or promotional strategies is much quicker than in previous years. 

You can always schedule it for later

Let’s be honest: the ability to schedule emails, ad campaigns, social posts, and even website updates has changed the way that many people work. When you finally make the decision to do something, you can schedule it for later, giving yourself that peace of mind that you can change it, even when you know you won’t need to. Bliss. 

Many people struggle to make decisions but ultimately, your inaction is less excusable than making the wrong decision, especially when a decision is backed up by date or an experimental hypothesis. 

Keep trying new things and leverage the plethora of resources that are available in marketing today to help you experiment and gather the data you need for better decision making.