At the heart of that change is the ongoing elimination of third-party cookies by the tech giants behind the world’s most popular web browsers, including Apple and Google.
These privacy-focused changes will make it more challenging for marketers to obtain the demographic and web tracking data they’ve long relied on, which is why there’s been so much pushback from the community.
But they also represent an opportunity to take a different approach and find new ways of doing things.
One of the most important steps marketers can take to adapt to this new reality is to embrace experimentation.
It’s something that’s long been critical to our operations at Uber, not least because we’ve been very careful to put data privacy policies in place which allow us to only capture data points that are critical to our operations.
So, for example, when we want to launch a new feature, we don’t do so anticipating that we’ll have a complete product straight off the bat.
Instead, we think about establishing what an ideal customer looks like and experiment with the product based on that. In some cases, this might entail launching slightly different versions of the product to different users, seeing how they interact with it, where they drop off and what brings them back.
This data can then be fed back into the product until it reaches a point where it enhances the customer experience.
This is the same process we follow in our growth strategies since it allows us to move fast, learn quickly and adapt according to what drives growth and the right client behaviour.
Market research still has value but over-reliance on it leaves you open to making inferences rather than data-based decisions.
However, if you want your marketing experiments to play that role, you have to know how to run them successfully.
The first mistake companies make is not to prioritise the build and design of technology that will allow them to run experiments seamlessly, across multiple teams (especially their growth and/or marketing teams), which allows them to analyse the results immediately.
Prioritise and invest in the necessary technology.
Then spend time formulating your hypothesis, clearly defining your primary and secondary KPIs, creating a controlled environment (as much as possible, but in the real world it will never be perfect and that is okay), thinking about execution before launch, and testing one hypothesis at a time.
Get this right and you’ll be well on your way to marketing that’s targeted and effective while still respecting the privacy of your customers.
