Saucey is a convenient alcohol delivery app developed by three friends over lunch. The founders wondered why there was delivery available for everything - groceries, household goods and even breakfast, but there was no app for alcohol delivery. As a company, Saucey wanted to enable all of their team members to spend more time analyzing data to make strategic decisions. Their data needs are robust, with customer facing applications, retail partners, and delivery logistics all needing to work together. Panoply allowed them to quickly set up a cloud data warehouse, with all of their data sources, and minimal data engineering support. And the love is mutual, we use Saucey for our office needs : )
Tell us about yourself and the history of Saucey.
I have a bit of a mixed background, I started off working in graphic and web design. Following that, I spent several years in post-production doing video editing and color grading. About 5 years ago I was hired onto the monetization team at textPlus, a social messaging application. For my first six months at textPlus, I primarily focused on building out and growing their earned and incentivized revenue. After that I became the iOS Product Owner and worked with an amazing group of POs and developers which helped form my love for product.
This is also where I met my two Saucey co-founders, Chris and Andrew. One day we were out at lunch and Chris commented on the fact that you can get pretty much anything delivered – groceries, food, household goods, laundry, etc. – but there was no existing delivery solution for alcohol, a use case that seemed perfect for delivery.
We started to look into the alcohol industry and pretty quickly became obsessed. It has an incredible history and is truly an old fashioned business that hasn’t changed much since the repeal of prohibition. There’s been slow evolution in the industry as a whole and we saw a large opportunity for us to add value.
We noticed in our own purchasing habits that we would not stock up on alcohol the way we would with food or household items. Instead we would pick up a bottle of wine, or a six-pack of beer when we wanted a beer or a glass of wine – and would almost always purchase from the nearest retailer or a store that was on our way home. This was much different from grocery shopping, since there are places I prefer to go to get my produce, meats, etc. When I buy a bottle of bourbon, it’s the same bourbon whether it’s from a local corner store or anyplace else. Regardless of minor shifts in cost or selection, we’d almost always go to the most convenient location. From this we thought the industry was largely driven by convenience and saw delivery as the most convenient option.
As we dove into consumer purchase habits for the industry, we found that our initial hypothesis was correct – that alcohol purchases are highly convenience driven and that the overall behavior was buy, consume, repeat. We learned that 80% of wine is consumed within hours of being purchased, and that it’s not uncommon for people that have a bottle or two of wine at home to save those bottles for special occasions, and still go to the corner store if they just feel like unwinding after a long day.
So we started building Saucey on nights and weekends. As we had a beta iOS app ready, we launched in West Hollywood, CA and soon expanded from there. What I don’t think we had fully realized at the time was that we would then spend the next year working on the product during the day and doing dispatch and deliveries in the evening. It was nearly a 24/7 job but we quickly became intimately familiar with all aspects of the business and those early days were invaluable.
What data needs does Saucey have?
There is a lot that goes into a customer placing an order and having it delivered in 30-minutes. We do our best to make it as simple as possible for the end user, and to do so there are a lot of components we need to pay attention to.
We essentially have three primary elements to our business, each with their own set of KPIs and targets to track to. We have the customer facing applications, where we track a lot of traditional e-commerce events and data – conversion, repeat and retention rates, purchase frequency, AOV, LTV, etc.
On the Retail partner side we analyze their menu to deliver as much consistency in offering and pricing to all of our customers in a given market. Additionally there are operational KPIs we hold our partners accountable for and work with them to deliver the data necessary to maintain a high level of operational consistency.
On the logistics side of the business we track to delivery times, fulfillment rates, ratings and are always monitoring and improving the efficiency of our batching and route optimization.
What were your requirements for BI/Data pipeline?
We need a solution set that was easy-to-use, reliable and fast. From a pipeline standpoint, we wanted plug-and-play while maintaining the ability to customize to our business’s specific needs. We didn’t want to internally worry about performance and therefore wanted a managed warehouse – and of course something that was cost effective as well. Panoply checked all of these boxes for us.
From a visualization standpoint – we also wanted a solution set that was easy-to-use, reliable and fast while maintaining flexibility. Something that a business user can use to pull reports independently, while still allowing for someone to run more complex queries and ad-hoc analyses when needed.
Tell us about your data stack:
Coming through Panoply we have Mongo for our internal data, Stripe for transactions, Twilio for messaging, Typeform for courier applicants and survey data, and Google Sheets for some one-off or offline data.
For visualization we use Looker and have been able to automate quite a lot. Employees have most of their everyday data readily available and have reports sitting in their inboxes daily or weekly as needed. It’s been great empowering everyone to pull and manipulate data on their own.
What was the original pain you were looking to solve by adopting a robust data stack?
Getting to a place where people can spend more time analyzing data to make decision, instead of pulling data or waiting on data requests. We think it is extremely important for employees to know our business inside and out, and therefore find it very important that they have clear insights into all aspects of the business, when they need it… not days or even hours later.
How do you use Panoply now?
Well, we’re as hands off as possible right now, which was the plan. The setup was pretty straight forward and because it works well, there isn’t much we need to do to maintain it.
We do still set up Redshift views in Panoply and the way you have set this up has made it very easy for us to add and manipulate views – in turn having performant, more complex queries that are readily accessible. The SQL workbench (the Analyze tab in Panoply) makes the management we do end up doing ourselves extremely easy.
Another factor I appreciate is the ability to organize and manage tables within Panoply. We do our best to keep things tidy so it is easy to navigate all the data we send through.
What’s a ‘big win’ Saucey has had with Panoply?
We’ve had trouble in the past properly transforming Mongo data with other ETLs, the way Panoply handles nested objects is perfect for us and solved what felt like a never ending issue.
We’ve seen drastic improvements in query speed as well on Panoply. Our Looker visualizations are way more efficient and fast now that we have everything hosted with Panoply.
How many people inside Saucey use the data warehouse/visualization tool?
Everyone in our company has dashboards and metrics they track. So, 100% of our company touches Panoply in one way or another.
Do you have anything to say in closing?
If you have data needs but don’t have the desire or resources to hire and manage a data team – Panoply is an amazing plug-and-play solution that doesn’t limit your ability to structure your data in the way that is best for your business.
