About BarkBox
Bark is a dog product company. It designs products and experiences for dogs and the people who love them, and it is best known for Barkbox, a direct-to-consumer subscription that ships monthly themed boxes of toys and treats. Bark also runs other subscription experiences and add-on products. More recently the company has leaned into a broader commerce strategy, connecting with major national retailers across the United States and expanding into marketplaces.
What do they do?
Industry
Location
The data landscape at Bark
Bark treats its data as core to how the business runs, and the way the team is built reflects that. A central data engineering team owns the platform from ingestion through to delivery, while a few departments keep their own analysts who still work out of the shared warehouse. Brandon Price, the Lead Data Engineer, sets the technical direction for the platform. Ingestion into BigQuery, the dbt models, delivery to the teams that depend on the data, and the platform's cost and performance all sit with him.
Because so much of the company touches data, the team pulls from nearly every corner of the business. Orders and subscriptions arrive from Shopify, down to customers, dogs, plans, and billing, plus the transactions behind its e-commerce and marketplace sales. Around that sit marketing and lifecycle signals such as channel spend and email and SMS engagement, customer feedback from support tickets, reviews, and NPS surveys, supply-chain detail across inventory and carriers, and the finance data behind the monthly close.
All of it lands in BigQuery, which the team runs as its single source of truth.
The cadence keeps the picture fresh. Source pipelines poll about every hour, and Looker refreshes twice a day, once in the morning so teams see the prior day and again later to show how the current one is tracking. Growth marketing, finance, and supply chain effectively live in those dashboards, and the executive team leans on them too. One pipeline goes further still. It reads NPS and review data and surfaces the themes customers are raising most, which gives leadership something concrete to hand to the product and customer-service teams.
How Bark modernized its data stack
When Brandon joined, Bark was running on an AWS / Redshift stack that he describes as fairly disjointed, with little in the way of a modern design. Pipelines fed into single tables that were also used for downstream reporting, so the same tables were constantly being written to and read from. That created recurring performance problems and expected downtime whenever the team needed to make changes. Many sources also landed without proper modeling, so there wasn't always a reliable source of truth.
The team took on the work of modernizing the stack and eventually moved onto BigQuery. That let them tackle the issues they had identified one by one. Introducing dbt let them model data correctly and set layered access. End users now see a clean, built-out single source of truth rather than the underlying tinkering, because their access is scoped to a governed warehouse layer.
The search for a reliable ingestion tool
Before Hevo, ingestion was a mix. Bark used CDP tooling for some data and a custom, REST-API-based pipeline for the rest, built to handle the load of an earlier period. As Shopify became central to the business and volume climbed, that custom pipeline started hitting data-freshness issues the team had to act on right away. Shopify data was now core to the business, so they needed a reliable product to own that ingestion, ideally with a backup on their own side.
Their first move was Fivetran. It handled the load at the time, but cost became the deciding issue. As Shopify volume grew month over month, Fivetran's costs climbed with it, to a point the team found untenable. Brandon points to a transparency gap. The trial estimate felt like a flat, optimistic projection that didn't reflect how quickly a live, growing business would scale. Once the team could see the real numbers, the tool was clearly too costly for them. Support during their Fivetran issues also fell short of what they needed.
Once you're in there and you can look at the numbers yourself, you can see it's a very costly tool. Month over month it just kept getting bigger and bigger, and we saw where it would end up. No, this isn't a tenable product for us anymore.
The switch to Hevo
Once Bark moved to Hevo, the transition was smooth and fast. After integrating Hevo's schemas into their models, the team was, in Brandon's words, off and running. From there the pipelines ran largely hands-off. Brandon rarely has to step in beyond the occasional field remap, and Hevo handled the growing Shopify scale.
It was very much a time-to-value gain for us. Once it was onboarded and we integrated the Hevo schemas into our model, we were off and running, a very smooth transition, without much engineering time after that.
The Hevo advantage in action
Asked for his top reasons to recommend Hevo, Brandon pointed to a few that map directly to the problems Bark set out to solve.
Predictable, respectable pricing: the pricing model fit what Bark was trying to do, and once understood it stayed hands-off. That answered the rising costs that had driven the team off their previous tool.
Hands-off operation: pipelines run with minimal intervention. The most Brandon typically needs to do is remap a field now and then, which frees the data team to focus on modeling instead of plumbing.
Responsive, actionable support: when issues came up, the team could get practical, timely help, a marked contrast to their earlier experience elsewhere.
Fast time to value: a smooth onboarding and quick integration into the existing dbt models meant Bark saw value without a heavy engineering investment.
There are very few times I have to go in and do something with Hevo. It's very hands-off. And the customer experience has been great; being able to get very actionable support was much appreciated
Excited to see Hevo in action and understand how a modern data stack can help your business grow? Sign up for our 14-day free trial or register for a personalized demo with our product expert.
