On-Demand Delivery, Instant Gratification… if you’re not used to hear these words already, you will be soon. The fast-changing space of on-demand economy is filled with startups that will bring anything to you, almost instantly.
Uber, Airbnb, Caviar, PushForPizza, Munchery, Doordash, Postmates, SpoonRocket, Sprig, Instacart, Shyp, TaskRabbit; all of them are great examples of this fascinating trend, and today we are interviewing Sergio Treviño, Co-Founder and Lead Technical Architect of BrewDrop.
BrewDrop delivers beer, wine and liquor to anyone’s home. How was the project born?
First of all, thanks for having me on your corner of the web. You are correct, BrewDrop delivers beer, wine and liquor on demand to our client’s desired location within the limits of a supported ZIP Code.
The project was born a little over a year ago when the Bell brothers in Austin were having a couple of beers at a bar and thought it would be great to have beer delivered to your home, quickly and without extreme fees. Also if you are running out of juice during a party, we’d keep our users from exposing themselves to the dangers of driving. Gerardo was brought in to set up a corporation and provide business and tech advice, and he has known me for a long time, so I was brought in to figure out a way to make the technology happen, focusing on a mobile-first approach.
Initially, the application would only deliver beer, but it was later decided to include liquor and wine, giving us the opportunity to capture a bigger market. The brand name stayed the same, reflecting the original idea for the service.
With the rise of the on-demand economy (ODE), soon there will be apps that will allow us to get almost anything delivered in a matter of minutes. Do you think ODE is here to stay?
Absolutely. It’s been there for a long time and it will be there for the foreseeable future. Pay-Per-View events have been available on cable since 1975 where boxing events were available for purchase. If you think about it, it’s been close to 40 years since this model started; people seem to be used to instant gratification, they like their things right now for cheap.
ODE + Mobile + Cloud Technologies have enabled a new wave of companies to tap into this market, including BrewDrop. Even though a parent company could offer services nationwide (or internationally), the impact of using these services is most directly felt on the local economy. Here’s the thing, to achieve instant and inexpensive deliveries we need to rely on local infrastructure, small businesses providing employment to their local communities thrive on the market’s demand for instant gratification. We get to serve our customers and businesses get an increased volume of sales. It’s not only a great option for consumers, it also stimulates local economies.
Does BrewDrop consider near-instant grocery delivery services like Instacart competitors at all? what’s your key differentiator?
There are many other near-instant delivery services. Blogs have repeatedly pointed out that we close the gap on Favor, for example, as they do not deliver alcohol. We are proud to say that we are the first mobile app that can offer these services in Texas (based on talks with the Texas alcohol regulation). There are other services taking orders by phone which are known to refuse orders at times because they are “too tired” to deliver. Not us, we are happily available during permitted business hours.
You talk about Instacart; BrewDrop, instead of going after a broad market like “all” groceries, pursued a narrow market, just alcohol. It is more effective for us as we get to know our target audience better. We can iterate faster and focus our service. That is our main differentiator.
Creating partnerships with providers and stores, delivering the goods as fast as possible… what’s the biggest challenge for companies that try to be “the Uber of *insert product here*”?
Standardization of process and compliance. We need to build a brand and be consistent to our users. This is hard to achieve when you distribute and rely on partners to execute on deliveries. I would say, there are 3 main challenges we face with a distributed delivery network: inventory, branding and delivery.
On inventory: we have about 1500 different products offered today. Complexity increases as you offer more products, and as we grow our business we will try to reduce the amount of offered products (according to demand data through metrics) so that we can offer the same experiences to our clients in Austin, Houston, Seattle or anywhere else. As we do this, we facilitate more businesses to comply with our offerings while at the same time keeping quality checks on the products.
This idea has been proven by other services like “Push for Pizza” in which you get two options, cheese or pepperoni, that’s it. It also creates this quick experience the user expects from near-instant deliveries. Compare to Seamless, where their offerings are overwhelming and their checkout process is somewhat convoluted. The user experience starts when they open the app, then continue to find and choose items, and this becomes easier and faster when there’s fewer but better options to select from.
On branding: showing one familiar face for the user, it doesn’t matter who the provider fulfilling the order is. As a company, we have allocated budget to tackle this issue in the immediate future. As you get more into the distributed model -in which anyone could participate offering services under the parent company- everyone should comply to standards. Lyft dresses cars with a pink mustache, Caviar makes couriers wear an orange t-shirt. Make the experience different from your competitors and make it consistent and easier to adapt by all service providers.
On delivery: it should not matter to the end-user which provider delivers as long as they identify it with your brand. We have been establishing automated checks on orders to make sure they get delivered on time. The idea is to automate most of it and just intervene when an exception happens. We have had many of those exceptions, be it that an item was out of stock or priced wrong. It should all follow a defined process that we can repeat in other locations. We have been working hard as time goes by and we learn from the market and business on how to improve them, pass them on to new providers and have a better user experience.
A huge issue that keeps other competitors out of the alcohol delivery business is different state rules regarding alcohol distribution, so this process may change as we enter new markets. In the end, all processes should help the business meet their goals on speedy deliveries. Also, all our providers have signed documents that define SLA’s.
BrewDrop connects users with local liquor stores. How do you control the delivery experience in terms of speed and friendliness?
As briefly discussed on the previous question, delivery is one of the key features. Fortunately until today, the providers have been excellent in keeping a good record when delivering order. With the use of metrics, we know the average times it is taking to deliver and we know where deliveries are in real-time. New logistics features will be introduced to help providers group their orders by arrival and delivery location to reduce travel times and transportation expenses. We have been careful to open new ZIP Codes until we know the same experience can be replicated either by the same provider or another that can cover certain area.
As for friendliness, the service is not perfect and there have been many suggestions, like the need to add special instructions field as well as visually identifying delivery people, a single point of contact for support, and many others, but we also have responsive services, apps and quick checkout process to help this point. Users go first in the world of on-demand economy.
It often looks like the biggest problem for the on-demand economy is regulation; Airbnb and Uber keep suffering when they expand to new locations. How does BrewDrop deal with this?
In Austin -our first supported location- we need to comply with TABC Rules. They are strict about alcohol deliveries. Believe it or not, BrewDrop has been reported to the TABC by local competitors (not real-time apps) claiming irregularities, yet we have been consistently able to work with them to make sure everything is in order.
One of the most important things we need to make sure of is that at delivery, the user provides a valid government issued ID, we even have to keep funds locked for that order until this part of the process is satisfied and verified. We are analyzing new potential markets and their regulations at the time.
In this data-driven world, are there any metrics you look at in order to get a sense of what users want or dislike?
Metrics provide an incredible insight into what the users want today, even about the things they will want in the future. They provide information as to what products are popular on a certain location or date. What day of the week are best for sales and which are slow, which will enable different offerings and promotions. Even if users don’t end up buying, the fact that a product was added to the cart or opened for details is valuable information. Maybe the product was too expensive? Maybe the imagery was bad? Maybe they didn’t like the description? Many new features in our roadmap depend heavily on analytics.
After a user completes a new order, what happens behind the scenes? How is the order passed to liquor stores? Are you running your own servers?
Awesome question, we get to geek out a little! I will give you a brief tour of how we do it behind the tiny screens (see what I did there?). We don’t run our own servers. It would be too costly to maintain our own infrastructure, paying for data, hosting, server updates, or even just by having someone (other than me) dedicated to the uptime. There are good services doing that right now and we leverage them.
Given that we deal with monetary transactions, we have secure servers keeping all of our users data safe. We have no way of knowing anything about their personal data, including addresses or credit cards. That is all stored on a PCI compliant service. All data transmitted is encrypted and sent over HTTPS, kept away from us on a dedicated safe service in the cloud. Business logic web services are coded in C# and hosted on Windows Azure from Microsoft. REST, JSON, you name it, all the usual stuff the cool kids do. Endpoints are super responsive as we are keeping lots of data in queryable objects in memory, reducing our disk I/O to a minimum as well as bringing SQL Azure costs massively down. All imagery is hosted on Azure Storage. Wouldn’t do it any other way (hosting wise).
Making the services extremely responsive was top priority. We designed the services to return only the most basic data we need from the services, keeping them lean, no bloating, showing respect for our users’ data plans. This, in coordination with using ListView controls on iOS and Android (using Appcelerator Titanium for native apps) which is by far the most performing control for displaying high amounts of data visually on the phone. While in testing phase, we noted that using regular formatted views would bring the app down to a crawl, especially when displaying many items like Whiskey (in which our main provider has over 50 different options).
Once the user is authenticated we return identifiable info to the user interface so they can select their preferred credit cards and addresses for delivery (card numbers are NEVER fully displayed again, ever, ever, ever). When the order is placed, they send us a token for their credit card and address -which changes all the time- for us to pass that info over to processing.
Payment processing server returns a status, if there are no issues, locks the funds and notifies the user that their order is being delivered by a provider and the timer starts running. They get an e-mail with a confirmation with the full contents of their order as well as support information. We exchange phone data between provider and client in case there are any issues at delivery.
At this point, we have sent many notifications and emails to our providers, we created a team feature in which several people can manage a provider location, including delivery personnel. Push notifications are sent to the team for them to manage orders through the mobile application once an order is placed. They confirm the order, assign it for delivery and the delivery person goes on their way to the location.
We have been working with providers to standardize the process so they can have their phone constantly reporting back their location. This is done through TCP Sockets and in-memory geolocation objects to keep track of them and report real-time back to the users. We will be releasing this feature on the next iteration both for iOS and Android users. Similar to Uber when they report all the cars on a map, except that you’ll see a tiny Beer Glass moving on the map. Real-time socket communications are a great way to enable awesome experiences for our users.
The last part of the experience is the actual delivery. The delivery person uses the app to confirm they have confirmed all valid identification at which point we change the order status to completed, we submit the funds for settlement and the order is complete. BrewDrop takes joy on facilitating the transaction and our customers delight on a nice selection of beer, wine or liquor.
What’s next for BrewDrop? Expanding to other cities? New features? Any plans of making a web version?
BrewDrop is analyzing new markets, including international ones. Making that jump would be great for us, and we have a very capable team working all the details. As for new features, yes: we are very excited to be working on the next wave of updates which will improve much on the experience for our users and introduce new features like collections of items, improved status reporting and more.
This represents an immense amount of work, and I’d like the team to be focused on the most important features. As an interesting fact, when we first released the iOS version, we had lots of people (not saying angry) excited to have an Android version, I can’t say the same for a Web version. We will stick to a mobile strategy for the time being to reduce our workload. If time and resources allow, we will explore the possibility.
We’d like to expand our service to other markets that don’t involve alcohol and for that, we have a few ideas brewing. Cheers!
Image via BrewDrop