Saltar al contenido principal

Make Your A/B Testing More Effective and Efficient

Anjali Mehra, Senior Director of Product Analytics at DocuSign, discusses the role of A/B testing in data experimentation and how it can impact an organization.
12 feb 2023

Photo of Anjali Mehra
Guest
Anjali Mehra
LinkedIn

Anjali Mehra is the Senior Director of Product Analytics, Data Science, Experimentation and Instrumentation at DocuSign. Throughout her career, she has also worked in marketing analytics and customer analytics, giving her significant insight into the impact of A/B testing and experimentation across all multiple parts of an organization.


Photo of Richie Cotton
Host
Richie Cotton

Richie helps individuals and organizations get better at using data and AI. He's been a data scientist since before it was called data science, and has written two books and created many DataCamp courses on the subject. He is a host of the DataFramed podcast, and runs DataCamp's webinar program.

Key quotes

Organizations should keep very close eyes on A/B tests while they are running. There are a lot of things that can go wrong, and teams make mistakes. It’s possible that you did something that's now introduced an error in your product or feature, but also it's possible that you influence a user behavior in a way that's really, really hurting you. For example, let’s say your test period is four weeks. Within just the first few days, if you aren’t monitoring your test, you may not notice that your order volume is going down significantly, and it will stay that way for the full duration of the test unless you catch it early. So, it's critical to monitor any test while it's running.

Whenever someone has a new idea, there is always a lot of excitement and urgency to jump into testing it right away, so spending some time in the planning phase may seem a little bit like a slowing you down, but it really does pay back dividends in the end. It’s important to invest time into  building a strong hypothesis because not having one just makes everything take longer and risks more since you can’t fully trust the integrity of the results. Both rushing the process and trying to measure against too many KPIs can derail any testing process and hurt overall business value.

Key takeaways

1

You need a strong hypothesis and a strong primary KPI to make A/B testing an effective and efficient use of time and resources.

2

A/B Testing is a team effort that requires thorough planning, active monitoring, and constant tweaking to get good results.

3

A cost-benefit analysis is a great way to identify if what you want to test whether or not the results you are aiming for are going to drive more value than the resources it would require to test and implement anything new.

Temas
Relacionado

blog

Data Demystified: What is A/B Testing?

In part seven of data demystified, we’ll break down one of the most common use cases of statistical analysis in data science: A/B testing.

Richie Cotton

10 min

blog

Data Science In The Trucking Industry (Transcript)

Discover how data science is being used to revolutionize the trucking industry, from A/B Testing and econometrics to machine learning and self-driving cars.
Hugo Bowne-Anderson's photo

Hugo Bowne-Anderson

48 min

podcast

Data Science and Online Experiments at Etsy

Discover how data science and online experiments impact business and decision making at Etsy.
Hugo Bowne-Anderson's photo

Hugo Bowne-Anderson

60 min

podcast

Scaling Experimentation at American Express with Amit Mondal, VP & Head of Digital Analytics & Experimentation at American Express

Adel and Amit explore the importance of experimentation at American Express, the pillars that feed into a culture of experimentation, frameworks for building successful experiments, robust experiment design and much more.
Adel Nehme's photo

Adel Nehme

39 min

podcast

How Salesforce Created a High-Impact Data Science Organization

Anjali Samani, Director of Data Science & Data Intelligence at Salesforce, joins the show to discuss what it takes to become a mature data organization and how to build an impactful, diverse data team.
Adel Nehme's photo

Adel Nehme

44 min

code-along

A/B Testing in R

Compare the performance of two groups with this introduction to A/B testing in R
Arne Warnke's photo

Arne Warnke

See MoreSee More