Alteryx Machine Learning recently introduced time series enhancements, and we are super proud of it! In addition to classification and regression, customers may now be able to perform predictions on time-based data over a period of time – this type of analysis is known as time series. From Alteryx customer to now Alteryx Advisor, read about @rachelhatcherian's experience with time series predictions and Alteryx Machine Learning.
What is your role at Alteryx?
I’m here to help customers on their data science and machine learning journey. So, that could be anything ranging from guiding them through their first steps to exploring use cases and talking through different methodologies that could be used or just helping to troubleshoot something that they're working on right now. Specifically, I help with anything ranging from the Designer Predictive Tools—sometimes called the brown tools—to Intelligence Suite to our cloud Machine Learning product.
Prior to working at Alteryx, what was your background with data science and machine learning?
Actually, funny enough, Alteryx is the reason that I got into data science. I was a data analyst, and I was using Alteryx to do my processes and reporting, which helped me automate around 80% of my job. And with that newfound time that I had on my hands, I started digging into Alteryx’s predictive tools. And then when Inspire 2018 came up, I ended up attending. They had analytic trainings, and I went to one on time series and some other ones like A/B testing as well. They were really good. When I got back from Inspire, I was able to implement the things that I had learned while I was there, and from then on, I was hooked. I went back to school, got my degree in data science, started moving into more data science-specific roles, and then eventually ended up here at Alteryx. So, it's come full circle for me, which has been funny and exciting. Now, as an Alteryx associate, I have been to Inspire several times including Inspire 2022 and Inspire EMEA (Europe, Middle East, Africa) 2022.
What is the new time series functionality?
So, time series lets you forecast into the future, but it's a lot more than that. Using Alteryx Machine Learning, you can not only forecast, but you can also perform exploratory data analysis along the way. This functionality allows you to look at things like trends and seasonality with time series decomposition. We can also look at things like the relationships in your data outliers. It's really a very interactive experience to not only explore your data but also to predict and forecast. And, what's great about it is that it really can be used by anyone. The focus is on upskilling and enablement and bringing in people that are new to data science.
The decomposition of time series deconstructs time series into several components that each represent underlying categories of patterns – namely trend, seasonal, and residual.
What makes you particularly excited about the new time series functionality?
My background is in finance and FP&A (Financial Planning & Analysis) and I really wish I had Alteryx Machine Learning when I was an analyst! What gets me excited about the time series feature is its ability to enable more people to perform what typically is a tricky process in an easy to use and guided way. Not only is Alteryx Machine Learning doing most of the heavy lifting for you, but something like our built-in education mode is really what makes this solution shine.
How have customers or other stakeholders been reacting to this feature?
There has been lots of excitement! I know a lot of customers were really excited to play around with the beta, and now that more features are out, there has been a lot of excitement and buzz around here, but they have been reacting specifically to the visuals, ease of use, and upskilling opportunities. A lot of leaders have been talking about the need to upskill their knowledge workers and analysts for the goals that they want to accomplish for the year. So, when we show them something like time series and the education mode, interactive lessons, and live trainings that come with Alteryx Machine Learning, it's a perfect fit. We can curate an upskilling and enablement plan for their organization, and that gets them excited, and it gets me excited too, because like I said, I love empowering people to do cool things like this. You don't have to already know a lot about data science to come into using Alteryx Machine Learning, because the barrier to entry makes it so that anyone can use it.
Is there anything else you would want to highlight to our readers?
Since Alteryx Machine Learning is a cloud solution, there are new features pushed out all the time. I'm constantly going in and finding new and exciting things within. And since our product is open-source, it’s not a black box, and this transparency is something that our customers really gravitate towards, especially in fields like finance.