A conversation with Bithika Mehra about Experimentation
Despite our best efforts, HiPPOs are key stakeholders that can make or break an optimization program. I recently spoke to Bithika about her thoughts on how to get HiPPOs on our side as well as potentially leveraging neuromarketing to improve conversion.
Rommil: Hi Bithika, how are you? Thanks for taking the time to chat today!
Bithika: I am very well, Rommil! Really excited to chat with you today.
Could you share with us a bit about yourself and a bit about your career journey thus far?
Sure! I pivoted into conversion optimization after spending a few years in digital analytics. I was introduced to testing when I joined Stamps.com back in 2015. We ran some major tests in customer onboarding, web registration and pricing that improved LTV and conversion rates. This was a part of my role at Stamps.com but over time it was one that I started enjoying the most. So, when the opportunity came around to build testing & optimization at an iconic brand like Fender, I jumped at it!
At Fender, we managed to improve conversion rates by testing in the registration flow and homepage. I also launched a user panel for Fender, which has become a key source of insights. We ran surveys and conducted usability tests for different teams within the organization.
As we were working on expanding the optimization function, CoVid-19 struck and the program was put on hold. Unfortunately, I was laid off following that.
Fast forward to now, I am freelancing and consulting for Fender and a few other organizations. I blog on topics in testing and optimization. Following Kelly Wortham’s (TLC founder) great advice, I also decided to “go back to school” and pursue the CXL mini degree in conversion optimization, which is very in-depth and highly recommended.
Connect with members of the Experiment Nation Directory
|Photo||Name||Location||Short Bio / Specialities||LinkedIn URL|
|Tracy Laranjo||Toronto, Ontario, Canada||Experimentation, Analytics, A/B Testing||https://www.linkedin.com/in/tracylaranjo/|
|Amrdeep Athwal||United Kingdom||CRO, web analytics and ux||https://www.linkedin.com/in/amrdeepathwal|
|Nadim Haddad||Toronto, Canada||Experimentation, personalization, analytics, BU reporting||http://www.linkedin.com/in/nadimhaddad/|
HiPPOs (or the highest-paid person’s opinion) often ask how an optimization or Experimentation program is performing. What are your thoughts about them asking how many tests were run? Is there anything that you would rather they ask?
I’d rather have a HiPPO ask me how many tests were run than not care about what I do. However, focusing on the number of tests being run could foster a culture of quantity over quality. That’s why I steer clear of any goals around volume of tests especially when the program is more mature.
“…I steer clear of any goals around volume of tests especially when the program is more mature.”
There are so many better questions to ask. I would definitely like them to ask what research and analyses are driving our hypotheses. How are we confirming our test results before implementing, iterating or moving on? What has been shipped out without being tested? What is the value that the optimization function is adding? Where do we see the biggest opportunities from optimization?
I love those questions. Those really speak volumes about whether an organization has truly embraced Experimentation. For places that haven’t quite come around, how do you suggest we partner with HiPPOs to change their way of thinking?
Hazjier Pourkhalkhali from Optimizely gave a great talk during #opticon20 where he shared some research being done on experimentation culture at organizations and the role of HiPPOs. The HiPPO not only acts as an executive sponsor for the experimentation program but is also critical for giving the team the autonomy to take risks and swing for the fences.
On the flip side, HiPPOs could kill that very culture by stifling the free flow of ideas and strongly holding on to their opinions even with conflicting data. As per Hazjier Pourkhalkhali, research shows that although HiPPOs have winning ideas, those ideas lead to a smaller lift than average. So, partnering with the HiPPO is critical for the success of the optimization program.
It is important to gain the HiPPO’s trust by speaking the same language. Growth in traffic and engagement means nothing unless tied to the metrics that are important to the business.
Testing also unlocks a lot of learning and yet, a lot of that stays scattered in different decks. So, it’s important to show the value of what we are doing by positioning the function as one that is augmenting the organization’s understanding of the customer and making everyone smarter.
At the end of the day, our role is to educate the HiPPO on the latest and the best in the field of optimization – whether it is third party research or what other companies have found success doing.
“…it’s important to show the value of what we are doing by positioning the function as one that is augmenting the organization’s understanding of the customer and making everyone smarter.”
You often hear some CROs talk about sure-fire tests and things you absolutely must test. What are your thoughts about these apparent best practices? What do you suggest Experimenters do instead?
There are no sure-fire tests as such. Just because something worked for some other org, it doesn’t mean the same thing will work for you. Testing based on best practices is fine if you have solved all identifiable customer problems or if you are just starting out and have little data to guide any meaningful testing.
Your customers are different. Their problems are different. So sure-fire tests may not even apply. And if you copy your competitors, then after a point it will be hard to differentiate yourself which Peep Laja calls being in the “sea of sameness”.
Instead, CROs should invest their time in identifying customer problems that are causing leaks in the funnel. Conducting a thorough heuristic analysis spanning technical and content itself could generate a list of potential improvements. Additionally, invest in improving the reliability of your data and tracking and conduct qualitative research such as surveys, user interviews and user testing to drive more hypotheses.
“CROs should invest their time in identifying customer problems that are causing leaks in the funnel.”
Changing gears a bit. You’ve recently written an interesting article on Neuromarketing. Could you tell us why neuromarketing is important to driving conversions?
It’s such a fascinating topic! A large majority of our decisions are made by the old brain. It is the most ancient and primitive part of our brain which we also share with reptiles. Less than 5% of our decision making occurs in the conscious rational brain. This means that as marketers we can be very rational and logical in our messaging but still not persuade customers to buy our product.
So, it is important to understand what appeals to the old brain. For example, the old brain is self-centered so a brand like Dollar Shave Club makes it about you rather than focusing on the ingredients in their grooming products or the convenience of the monthly subscription.
For technical or B2B products, there could be pages on your website with a ton of text which the user could ignore. However, we know that the old brain is triggered by change. Finding changes in patterns – predator emerging from the bushes – used to be about life and death. So disrupting the pattern by using design and layout can help keep the user’s attention.
So there are lots of ways that neuromarketing principles can be applied to marketing.
Having worked with a non-profit recently, how would you suggest that non-profits leverage neuromarketing?
There are a lot of different ways non-profits can use neuromarketing principles to drive say, donations. Emotion is a big one. It triggers the old brain and you typically do see a lot of emotions-heavy messaging from non-profits generating empathy, fear etc. In fact, we are running a test on Greenpeace Africa’s website where we added copy and visual in the donations flow to drive motivation to donate.
There are other ways. Identity is important. Local public radio stations have an exclusive leadership circle for the highest donors. Social proof by way of testimonials from recipients is another one.
Very interesting stuff. I definitely encourage our readers to check your post out.
Finally, it’s time for the Lightning Round!
Frequentist or Bayesian?
Either one if testing is conducted with full rigour and adequate sample sizes.
Do you own a Fender?
Yes, a beautiful red Fender Telecaster Thinline!
If you couldn’t work in Experimentation, what would you do?
I’d be a veterinarian!
Describe Bithika in 5 words or less.
Driven, buoyant, empathetic
Thank you, Bithika, for joining the conversation!
Connect with Experimenters from around the world
We’ll highlight our latest members throughout our site, shout them out on LinkedIn, and for those who are interested, include them in an upcoming profile feature on our site.