CouponFollow: An Active Metadata Pioneer


Driving Data Discovery and Improving Context with Atlan

The Active Metadata Pioneers series features Atlan customers who have recently completed a thorough evaluation of the Active Metadata Management market. Paying forward what you’ve learned to the next data leader is the true spirit of the Atlan community! So they’re here to share their hard-earned perspective on an evolving market, what makes up their modern data stack, innovative use cases for metadata, and more.

In this installment of the series, we meet Ted Andersson, Director of Business Intelligence at CouponFollow, who shares how CouponFollow’s significant business growth has driven similar growth in their data, how he and his team evaluated the Active Metadata Management market, and how Atlan can serve as the “glue” between their data practitioners and consumers.

This interview has been edited for brevity and clarity.


Could you tell us a bit about yourself, your background, and what drew you to Data & Analytics?

I have an academic background in mathematics and statistics, graduating with a Master’s in Statistics. I have been interested in data for as long as I can remember, so it was a natural thing for me to go from working in the statistics realm to working in the data realm. 

My career in data started at a small influencer marketing company, where I was first introduced to affiliate marketing and the tech world. From there, my journey has taken me through various industries and roles. I’ve worked on projects ranging from call volume projections and optimal staffing levels at a tax and accounting firm, to developing a chatbot for customer support questions. I also had a stint at an EMS services company, helping them transition their software from on-premise to the cloud.

I returned to Affiliate by joining Rakuten Rewards, and here I am at CouponFollow, where I’m the Director of Business Intelligence, responsible for anything Data. Anywhere there’s any data that’s created, modified, transformed, or visualized at CouponFollow, my team is responsible for it. That team encompasses Data Engineering, traditional analysis, and even the early stages of pushing into Machine Learning and Data Science projects.

Would you mind describing CouponFollow?

CouponFollow has traditionally been a coupon and deal site. Our founder is an expert in SEO and he’d been growing CouponFollow.com for more than 10 years, but with the pandemic there was a massive surge in interest in coupons and online shopping. 

As the company continued to mature, CouponFollow started to bring in a much larger team, including me, who was brought in to start the data team. At the time, they were working with a consultancy that had gotten all the “plumbing” in place, but they needed someone to help set the vision and strategy for data at CouponFollow.

The most important part of our business is organic search. When someone goes to Google and types in something like “Macy’s Coupon,” we want to be as high as possible in that ranking. We use Affiliate to monetize the click and that’s the most important part of our business. 

That being said, we’re at a place now where we’re starting to diversify, moving into the whitelabeling space with an offering for CouponFollow to build branded shopping experiences for select partners. Customers for this offering might be news or content sites, where you see coupons and shopping as part of their offering. 

We have a few different strategies, trying to scale a business away from just being focused exclusively on organic search, toward more opportunities.

What does your data stack look like?

We have Snowflake, dbt, and we use Sigma for visualization. CouponFollow was acquired about a year ago by a larger company called System1, and they’re a Tableau shop, so we use a little Tableau to accommodate their needs, and use Sigma for our own reporting. Beyond that, we use Snowplow for real-time events, and a tool called Indicative for visualizing how our traffic is doing in real time.

Why search for an Active Metadata Management solution? What was missing?

This started off when I asked my team to document dashboards and data models, and they would create something in Confluence. Number one, nobody looked at it, and number two, by the time it was written, it seemed like it was already out-of-date. So the approach was just not scalable. We would build a dashboard, then document it, but as soon as someone changed that dashboard, the documentation would be out-of-date. When somebody would go to get an answer, the documentation was not giving them what they needed. 

Additionally, definitions themselves can be all over the place. One person might think of Click-through Rate (CTR) to mean one thing, and a different team thinks of it as something else, and they each have their own dashboards that have CTR, but they aren’t aligned.

When these teams aren’t aligned, they ask “What’s the BI Team doing?” or “Why are they saying that” and you end up with lots of questions and confusion. CouponFollow ingests a tremendous amount of data from all kinds of different places, and it’s very challenging to keep track of all of it.

Because of all of this, a tool like Atlan really became top-of-mind.

Why was Atlan a good fit? Did anything stand out during your evaluation process?

We began when I talked to somebody who was working with Atlan and recommended it. Because of that, Atlan was the starting point in our search, but we always like to do our due diligence. 

So I did some shopping around to look at the other options, like Alation, Castor, and Collibra, and had conversations with all these companies. Some of them certainly seem like they have good tools, but they might be focused just on very large organizations. That’s not us, and that’s not what we were looking for. So, it came down to Atlan and Castor.

Castor offered a discounted rate below what Atlan was charging. But I think what pushed me and my team over the edge for Atlan was it just seems like Atlan is the more comprehensive, state-of-the-art tool, and that Castor is playing catch-up. They just seem like they’re a little bit behind Atlan. 

And I liked the sales approach of Atlan. Generally speaking, I didn’t find the team to be pushy or “salesy”. It was more like “This is our tool. We know what we’re about, we stand behind what we have, and we want to work with you as a partner.” 

Sometimes we’ll get these weird sales pitches talking about features that are coming in the pipeline, maybe in six months, maybe never, saying, “Well if you go with us we’ll be good in two years.” Or there’s throwing shade on the competition.

The Atlan pitch wasn’t about that. I found Atlan to be very direct, straightforward, and that you are what you are and you’re confident in your product. I like working with companies like that.

What do you intend on creating with Atlan? Do you have an idea of what use cases you’ll build, and the value you’ll drive?

My team is in the planning stages of a pretty big reorganization. The role of data at CouponFollow has grown tremendously from what was originally called the Business Intelligence team, but we’re really a Data Engineering team. 

We started with pulling data into the warehouse, and building some limited visualization and reporting. Now we’re in a place where we’ve gotten traction, and there’s a lot of different directions to go in, a lot of data, and a lot of scale. So the number one priority is getting this reorganization done so we can account for all these new, different processes we’re following with our data.

I want Atlan to be the glue that pulls all of this together, and provide a single, shared place where everybody can go to understand the data, how it flows, and where it comes from. Our first three months involve rolling Atlan out to my own team, and not too much end user or business stakeholder involvement yet. We’ll use it day to day, understand what’s in Atlan, and then we’ll do limited rollouts to specific stakeholders who are a bit more data-savvy to get them engaged.

Down the road, I’d like to see our business users empowered to use Atlan as a sort of “social place” where they can come in and make a comment if something looks weird. If they have questions like “What’s the source?” or “Who’s the owner?” they can do their own exploration, and they’ll know where data is coming from, and which person to talk to. 

Right now, if we don’t give business users a link to the dashboard they’re looking for, it’s very hard for them to find something on their own. Atlan will be a great place for them to search for something like the most popular dashboards so they can understand how they could use them. That’s the kind of behavior I’d like to see more of.

Did we miss anything?

For the purposes of anyone reading that’s trying to find the right tool, I would say the differentiating factor with Atlan is its position in the market. It’s this “sweet spot” where I think Atlan can work with any company, whether it’s a company that’s our size or an early stage startup, as long as they have somewhat of a data stack. And I can see it growing to support someone like Procter & Gamble, as well.

I think Atlan can grow with our company as it grows, too. The product is well built-out, and any features I would like to have seen in my wishlist were already there. There wasn’t anything lacking where we felt we would have to buy early, then sit and wait for features to come online.

So I think there’s a unique value proposition that the competition can’t quite match. Even with their discounts, it was sort of a no-brainer to go with Atlan.

Photo by Julio Lopez on Unsplash

Latest articles

spot_imgspot_img

Related articles

Leave a reply

Please enter your comment!
Please enter your name here

spot_imgspot_img