5 CRO Takeaways from NUX5


Myself and Josh from the Banc CRO team went to the fifth annual NUX conference here in Manchester last Friday. Here’s our five takeaways from the event, complete with obligatory blurry shots of the presenters’ slides to prove we were there.

 1. It doesn’t matter how much lipstick you put on a pig. It’ll still be a pig. (Henny Swan)

NUX CRO takeaways

In CRO it’s often quicker and easier to put a hack in place than address the root cause of a pain point.

A classic example is in a checkout process where certain form fields aren’t required to fulfil the order but can’t simply be removed from the form because of dependencies in legacy processing systems. In this instance a hack might be to use in-line validation or tool tips to make completing the fields easier, even though you ultimately don’t need the information.

While in the short term the hack can make it slightly simpler for your visitors to get through your website journey, not addressing the root cause can be a lot more limiting or destructive commercially – not to mention that further building on top of root problems can make them even harder to fix.

For the best long-term results, don’t ignore the underlying issue.

2. Your customers’ perceptions of you constantly change – and not because of anything you’re doing. (Sophie Dennis & Graham Odds)

NUX CRO takeaways

Sophie Dennis spoke about “expectation escalation” that stems from the way the likes of Amazon, Google and Uber continue to change every element of our world.

Such as how the fact that I can sign up to Uber and get a taxi outside my house all within 5 minutes changes my expectations in other areas of my life, like how long it should take to receive my shopping delivery, or how much time it takes my bank to process my credit card application.

Right now, transformative change is happening faster than ever – and the companies that succeed long-term are the ones who understand and embrace the future.

In his presentation Graham Odds spoke about how AI chatbots may soon transform the way we bank and manage our interactions with banks, retailers and almost every organisation we interact with.

That’s not to say we now all need to drop what we’re doing now and hire a chatbot team, but we need to be aware of how that type of growing technology might change people’s perceptions of how they should browse, purchase and interact with companies, so that we can adapt to meet those expectations.

3. …but just because something’s new doesn’t necessarily mean it’s better. (Boon Sheridan)


Boon Sheridan’s family suspected their cat was going deaf, so his father mentioned their concerns to the vet while booking the cat’s next check-up. The vet replied that the surgery would do a cat hearing test at the end of the appointment to make a diagnosis.

Boon’s father then researched exactly how cat hearing tests were conducted, and it turned out that there was a cool new technology being used in the area. Vets would position tiny modules inside the cat’s ears which would then detect the micro movements indicative of hearing activity.

Boon’s father was very excited to see this in action.

The day of the cat’s check-up arrived, and towards the end of the session, the vet told Boon’s father it was time for the hearing test.

Boon’s father was beside himself. This was going to be amazing.

And then the vet stood directly behind the cat and clapped loudly behind its head. The cat didn’t react. The vet: “Yep, your cat’s deaf – that’ll be $75.”

The moral of Boon’s story is that sometimes the traditional ways are just as good or better than the transformative new ways. The questions to ask when assessing a new piece of software are “Do we have the resources to use this effectively?” and “Will it improve our bottom line?”

4. Everyone else has the answers. (Karina van Schaardenburg)

NUX CRO takeaways

Researcher Karina van Schaardenburg spoke about her time studying Foursquare users based in Turkey, and the blend of qualitative and quantitative methods she used to get inside the heads of that audience.

Her methods mirror those we use in CRO. Your segmentation analytics may reveal the majority of your tablet visitors are from Manchester, but that your London tablet users are most twice as likely to convert. Realistically, only qualitative research like user interviews or focus groups can tell you why, but you would never identify that issue in the first place without a deep dive into the quantitative data.

Similarly a problem spotted by one person can be validated and prioritised for a fix based on how common it appears to be based on your site data.

CRO becomes a lot easier when you realise that, 99% of the time, everyone else has the right answers. The website users interacting with your pages or AB test, or leaving feedback on your site. The call centre or store staff you speak to about common customer problems they come across. The brand or marketing teams running user research groups to understand what your company voice sounds like.

As CROs, all we need to do is ask the right questions, and then listen.

5. Never assume something will work just because it’s ‘normal’. (Boon Sheridan)

NUX CRO takeaways

Boon Sheridan was talking specifically about UX design when he said this, but the sentiment is equally true in CRO.

Just because something used to work (or is used on a different website) doesn’t mean it will work for you now. Things change and every business is different anyway.

To create the best website performance, we need to constantly learn about our users and their behaviour. There are no shortcuts to be had by only copying competitors or best practices. Even if we get lucky and see uplift from a site change we copied from Amazon, what have we learnt to help us next time we’re looking to improve something?

Real conversion optimisation takes time, and it’s a continual process that gets stronger the more we learn about our customers. If we make it a numbers game where the main aim is simply to run as many tests as possible, we compromise on results and end up with bad data and no insights.

A CRO does a good job over the course of a quarter, a year or more. Not over the course of a couple of tests.

So there are our five CRO takeaways from NUX5. Roll on NUX6, where we’ll have to find an additional sixth takeaway to keep up the consistent numbering. If you’re interested in talking to us about CRO, get in touch at cro@bancmedia.com.

Avatar for Anita Pun
Anita Pun