A Conversion Conversation with IBM’s Daniel Martin del Campo
I had the chance to talk to Daniel recently, and he shared with me his thoughts on what it takes to succeed at Experimentation, the biggest mistake people make in Personalization, and what advice he’d give his younger self.
Rommil: Hey Daniel, nice to meet you! Let’s start off with you sharing a bit about yourself and what you do. Could you tell me a bit about how you got into experimentation?
Daniel: I started as Web Developer but I have always had interest in design and user experience, I’d strive to create visually appealing sites that feature user-friendly design and clear navigation and often got request to A/B test some design and content on our pages. I was always curious about the strategy behind it, so I’d investigate and learn more about the experimentation strategy.
What key skills should someone who’s interested in getting into experimentation have?
I think curiosity is a must in experimentation. Experimentation involves asking a lot of questions and curiosity is the fuel that will get you to ask those questions, wether it is from “what if we change this CTA color to green?” to “how does this page drive value to our users?”. Curiosity is something you must have if you are interested in getting into experimentation.
Analytical and problem-solving skills. The ability to analyze your users data prior to launching an experiment will help you better understand your users behavior and needs, it will allow you to ideate a stronger hypothesis. Being analytical and a problem solver is specially great skills to have if you are getting into personalization, because by understanding your audience problems you can then come up with a personalization campaign to solve those problem.
Here’s a tougher question — what does a company culture of experimentation mean to you?
Interesting question, I would say working in an environment where everyone is in sync with experimentation and understands its value, a place where everything revolves around experiments and where everyone is exited about learning and even if your teams ideas don’t turn out to be the winners it doesn’t matter because your team has learn. An environment where everyones knows that experimentation can drive innovation.
Why do you think some companies have a hard time embracing a culture of experimentation?
It could be for several reasons, but what I think is the number one thing why some companies have a hard time embracing a culture of experimentation is because they don’t see or haven’t discovered the value that experimentation can have in their business. It might be because not everyone is bough in on experimentation therefore they dont have experimentation as a priority.
If you could name 3 things that companies should do to embrace experimentation as well as to see the rewards from it, what would they be?
Connect your business metrics to your experiment metrics, by doing this you will notice the impact that experimentation can have in your business. Enablement is a big one also, as your team gets better and better at ideating and creating experiments so does the demand for everyone to ideate and launch more complex experiment, try to host monthly experimentation office hours and have central teams dedicated to experimentation. And last but not least, invest in necessary tool that will facilitate experimentation.
If you could go back in time, what advice would you give your younger experimentation self?
I consider myself to be fairly new in experimentation I’ve been in this role for roughly less than a year. So if had to give advise to me current self, I’d say don’t get frustrated if your ideas don’t result on a winning variations. One of the things that I have learned so far is that experimentation is all about learning and when you have mindset oriented to learning, you always win.
In general. I would say, don’t be afraid to ask questions, when I was younger I would listen to people talk and if I had doubts or questions about what they were saying I would just stay quiet and would assume that I understood.
What do you do as a web personalization strategist?
As a Web Personalization Strategist my purpose is to create personalized experiences for our users. I work for IBM Latin America and one of the things that I do is align IBM’s global Personalization strategy to the Latin America markets. I work cross business units and work closely with key stakeholders to Ideate and launch personalization campaigns. My goal is to Integrate a successful personalization lifecycle framework, integrate tools, templates and different communication, collaboration and documentation practices so all personalization campaigns run as smoothly as possible.
How do you experiment on personalization?
This is a good question. When you create a personalized experience you are basically experimenting with a specific audience out of all your users. When you set up your campaign, you identify your targeted audience and instead of allocating your traffic to 50/50 you allocate it to 5/95. Where 5% of your audience will see the original variation and 95% will see the personalized variation, it will take longer to reach statistical significance compared to an A/B Test, but you will eventually get there. One way to approach this is by creating an “always-on” personalization campaign. What I mean by “always-on” is creating personalization campaign that can be iterated upon. For example, if you personalized experience is underperforming, you can analyze your results and understand why is underperforming so when you launch your next iteration you will have a better understanding of what your audience needs and will target them with better content. The idea is to always learn from your audiences and get better with each iteration.
What’s the biggest mistake people make in personalization?
I would say not understanding your audience or creating a personalized experience for the wrong audience. This isn’t entirely everyones mistakes, this can sometimes be cause by a lack of data and information about your audience behavior and not having the right tools to leverage this data.
Finally, Bayesian or Frequentist?
Awesome! Daniel, thank you!
You may also like