When it comes to marketing your product, there are a million different ways to do it.
But how do you know which method is the right one for your brand?
The answer is marketing experimentation!
Regular experimentation ensures you stay on top of new trends, audience changes, and fluctuating data.
With the right experimentation programs in place, you and your team can pinpoint what works and what doesn’t, helping you to make strategic changes that earn more sales.
In this blog, we’re covering all things marketing experimentation. Plus, we’re sharing hard-earned insights from Alice de Courcy, the CMO here at Cognism.
Lab coats and goggles on, please! 👇
What is marketing experimentation?
Marketing experimentation is a systematic approach to testing and optimising your marketing performance.
If you’ve ever sat through a school science class, you know that experiments start with a hypothesis and include variables to test what happens if something else occurs.
Marketing experimentation works the same way. In other words, you test your marketing strategies to determine which yields the best results. As you test, you change one variable at a time to accurately measure the impact.
When you manipulate certain variables, like spending more or less on ad spend or cold calling at different times of the day, you learn what works best for your audience and your brand.
What are some examples of B2B marketing experiments?
There are plenty of interesting experiments you and your team can run. For example, you can run tests like:
- A/B testing on website pages.
- Multivariate testing.
- Price sensitivity testing.
- CTA testing.
- Landing page optimisation.
- Email subject line testing.
This list of experiments is just the tip of the iceberg. There are plenty of ways you can test your strategies and optimise them.
And the number one benefit of experiments in marketing?
It can save you and your team time and resources, especially if you’re on a limited budget.
How did Cognism run marketing experiments?
Here’s an example from Alice de Courcy’s Diary of a First-Time CMO:
When Alice stepped into the CMO role, resources were limited. This meant she needed to get creative with her experiments. Instead of spending hours creating assets the Cognism audience might not be interested in, she tested the waters first.
She said:
“The product team operated a very lean, MVP-first approach and I had very minimal resources to work with. So I’d adopted this way of thinking and found it to be a useful approach to getting things done and start learning quickly. So when it came to experimenting at Cognism, I followed this same philosophy.”
“For example, in the old lead gen days, before creating a long-form content asset, I would often launch an early bird sign-up landing page to see how much traction and demand there was for the asset. It was a great way to validate the topic and idea before investing weeks in writing and creating it.”
This method, called smoke testing, helped save the team time and effort. Ultimately, it yielded better results.
Why do B2B marketers need to experiment?
Marketing experimentation isn’t a one-and-done kind of thing. Instead, it’s a mindset and a continuous learning opportunity.
The more you understand your potential customers, the better you can adapt your strategies to meet their needs. And the better you are at adaptation, the more revenue you earn for your brand.
It’s helpful to tie some of your marketing KPIs to your overall revenue goals. Although the marketing department might not own revenue operations, marketing certainly helps drive brand awareness, which, in turn, leads to revenue growth.
As Alice explained:
“When I did take the leap and started setting goals and KPIs against revenue, everything else became a lot easier. Weirdly, the fear of missing target reduced.”
“I felt like suddenly I had more freedom to test out the tactics I’d been dying to try that weren’t directly measurable - because as long as they drive the end result, i.e. revenue generation, then that’s all that matters. I think you only get that freedom by signing up to a revenue target.”
The more you experiment with your marketing, the more you’ll likely find the natural jump from lead generation to demand generation.
Instead of looking for short-term wins, you’ll focus more on building brand awareness and authority and nurturing long-term client relationships.
As Alice said, as long as your strategies drive growth, that’s what matters.
What are the steps for running B2B marketing experiments?
Let’s break down what it takes to create a successful experiment.
1. Identify objectives
Remember what your science teacher told you:
You must understand your baselines and plan your objectives before doing anything else.
The same applies to running experiments in B2B marketing. You and your team must identify your objectives, goals, and metrics, especially if you want your experiment to be a success.
Take Alice’s approach. She said
“We built out a clear process around each type of content, including goals and objectives, depending on the individual use case and an understanding of what it would mean for each to be successful.”
Clearly defining your goals and objectives, plus understanding your success metrics, can help you tweak your strategies later on.
2. Develop hypotheses
With your list of objectives, it’s time to play the “What If?” game.
Like:
“What happens if we focus our efforts on creating more thought leadership and less top-of-the-funnel content?”
Or:
“What happens if we gate this content? Will it drive more leads?”
By asking, “What if?”, you can develop a hypothesis.
For proper results, you’ll want to formulate a testable assumption based on prior insights.
As Alice said, it’s “no different to how you should approach any new and untested activity.”
As you develop your hypothesis, consider your budget. Look at what Alice said as an example:
“I requested a 5K/per month ‘CMO testing budget’, then created a proposal detailing how I’d use this budget for 4 months to run ungated, demand gen plays using our top-performing content.”
“This involved optimising for consumption and not for conversions. My hypothesis was that I would start to see an increase in inbound demo leads in line with this additional spend on demand gen.”
If you don’t have the budget to play with experiments, don’t worry. There are plenty of “free” ways to test your processes. More on this later!
3. Design the experiment
As I mentioned earlier, there are a ton of tests you can run. However, the trick is picking the proper test for your hypothesis.
For example:
If you want to understand the customer journey, running an A/B test on CTAs might not give you the results you’re after.
If you choose the wrong test, you risk measuring variables that have nothing to do with your hypothesis, which can cause you to lose time, resources, and, ultimately, revenue.
This is why it’s imperative that you analyse your current metrics before designing an experiment. You need to have a solid understanding of where your brand is, where you want to go, and what you want to test.
Alice had some valuable insights here. She said:
“You can take ideas from what we have done and test them for yourselves, but realistically, without adapting them to what fits your business - it would never work the same way for you as it did us, because every company is different.”
“So it comes down to finding a process you can deploy each and every time you need to find a solution to a problem or start up a new marketing program.”
“One that allows you to get under the hood of your organisation, understand what is happening now, what you want for your future state and what needs to change in order for you to get there.”
Pro tip:
Take the time to research various marketing experiments. Look into the “how” and “why” of each test. This ensures you choose a test that provides results that speak to your hypothesis.
4. Execute the experiment
Once you’ve found the appropriate test, it’s time to run the experiment.
The most crucial part of executing an experiment is ensuring you implement it in a controlled manner. This means keeping your control variable constant while testing individual variables one at a time.
Let’s look at an example:
Alice wanted to test how best to create organic LinkedIn ads that people actually wanted to read and interact with. To do this, she made an organic LinkedIn post and then turned it into a promoted ad.
She said this about the experiment:
“We decided to test out this approach because we want to engage as much as we can with people in-feed, and our organic posts were performing much better than our ads. We also felt this would be a great way to get more information in front of our target audience.”
“From this, we decided our next best step would be to test a bespoke retargeting funnel for those people who have seen our influencer ad, in an attempt to create an aligned, personalised journey.”
“If we don’t see an uplift in results after testing this method in the coming couple of weeks, then we wouldn’t continue it and we’d choose something else to test instead.”
“We’re not going to worry about the problems around scaling it until we know it’s something that works. Because getting it live and executing something is better than doing nothing.”
Pro tip:
Executing something, whether it’s a “perfect test” or not, is better than doing nothing at all. Give your experiment time to yield results before testing a different variable.
5. Analyse results and implement findings
The best part of marketing experimentation is that it uncovers the truth behind your efforts. Here, you’ll learn what’s working for your brand and what you should change.
Ideally, your test yielded results that help you make more informed decisions going forwards. Use what you’ve learned to adjust your campaigns (or scrap them completely and try something new).
Remember, experimentation isn’t a one-and-done kind of thing but a mindset. Make it a habit to keep testing your strategies.
The more you test, the more you’ll learn. And the more you learn, the more opportunities you have to generate revenue.
What are the common challenges and solutions in marketing experimentation?
Experimenting with your marketing is necessary. However, you’ll likely run into some challenges, too.
Before you start experimenting, let’s look at some examples of common roadblocks and solutions to overcome them.
Limited resources
Earlier, Alice mentioned working with a small budget. Although a small budget can be challenging, it doesn’t have to be an obstacle that holds you back.
Instead, look for ways to optimise it. That might mean bringing your entire team together to brainstorm new and creative ideas.
Alice did this at Cognism. She said
“I love getting the whole team together to do some brainstorming; we come up with such great ideas when we have everyone bouncing off one another. This was an H2 planning workshop where we made some choices for how we were going to work moving forwards. One decision was to give our writers more creative freedom.”
The “creative freedom” bit is important. It was an easy way to optimise the resources Alice’s team was already working with. Giving the Cognism writers more freedom enabled them to experiment without spending more money to try something new.
Pro tip:
Look at your existing processes and make “free” changes before increasing or spending your budget. Not every test needs bags of dough behind it to make a difference!
Data reliability
A good marketing experiment relies on good data!
You’ll need accurate information to help plan your experiments before you can begin to think about execution.
Alice said:
“During a recent audit of team OKRs, we found that each team was storing their OKRs (objectives and key results) in different places, leading to a lack of visibility and alignment across the department.”
“This fragmented approach made it difficult to ensure that all teams were aligned with our common goals. Many OKRs were overly task-focused, which made it challenging to measure outcomes effectively or understand the broader impact on company objectives.”
“This ‘set it and forget it’ approach increased the chances of misaligned projects and unmet goals.”
“To fix this, we shifted to an outcome-focused OKR model. By defining clear, results-oriented goals and metrics, we’re now better aligned with the company’s broader objectives and can measure our impact more effectively.”
Pro tip:
Create clear, defined metrics and goals.
Once you’ve done that, ensure your team knows where to store and access data. This will keep everyone on the same page and ensure you don’t miss anything important.
Organisational resistance
It can be easy to forget that your sales and marketing teams play for the same side.
When refining your marketing strategies, you might find the CEO or CRO has questions. Or they might resist trying something new, especially if you can’t guarantee positive results.
At Cognism, Alice found that meeting with the CRO to combine sales and marketing efforts was the best approach, especially since they had the same goal: revenue.
To merge the teams, Alice and the heads of other departments have dedicated meetings to keep everyone up to date.
Alice said:
“We meet consistently to have sessions where we…
- Review our goals.
- Work out where we are in relation to meeting these goals.
- Decide what else we can do to reach where we want to be.”
Pro tip:
Transparency usually overcomes organisational resistance.
Keep your key stakeholders informed about your experiments, what you’re testing and what results you expect to see, especially in terms of marketing ROI or conversion rates.
How do you measure the success of marketing experiments?
This might sound a bit out there, but success isn’t always measured by numbers.
Sure, your metrics can be an indicator of success. And it’s always good to track things like the number of sign-ups you have on a webinar or the number of leads that convert from a paid ad.
However, success comes in various forms. For example:
When Alice and her team decided to host webinars, they worried less about sign-ups and more about the webinar content’s quality.
Alice said:
“Success would now be measured on how many great quality snippets we could create from the webinar to use on paid and organic socials.
- How many organic LinkedIn posts could be taken from the webinar?
- How engaging were the chat and questions during the live event?
It’s amazing how when I changed the KPIs for the team, how much better in quality the webinars became. With that, the attendees and sign-ups followed naturally.”
Pro tip:
Don’t be afraid to be creative with your metrics!
Non-numerical metrics, like customer feedback or the quality of participant interactions, can meaningfully drive your marketing strategies.
Get more actionable insights
Marketing experimentation isn’t a strategy - it’s a growth mindset.
The ability to test, learn, and iterate ensures your marketing stays agile and data-driven. Plus, the more flexibility you have within your marketing, the better you can connect with your audience.
As Alice’s insights prove, successful marketing doesn’t follow a strict playbook. The best B2B marketers use old ways of thinking as stepping stones to refining their strategies - and they do it all through trial and error.
The key is to start small. Follow these key steps:
- Define a hypothesis and select your experiment metrics.
- Test strategically, changing one independent variable at a time.
- When the test is over, use the findings to guide your next steps.
By embedding this culture of experimentation into your marketing, you’ll set your brand up for long-term success.
Want more of Alice’s insights? Dive into her real-world experiences and strategies running marketing at one of the UK’s fastest-growing scale-ups. Read her Diary of a First-Time CMO today! 👇
