--- title: Why Astronomer Support Is Chasing Michelin Star Service description: >- Astronomer’s Customer Reliability Engineering team was recently honored to receive an award by Business Intelligence Group for our support, a step toward our audacious goal of being the best software support team in the world. We are proud to work with such an incredible customer base, many of whom use Astro for mission-critical workloads. That responsibility demands an uncommon level of service and excellence. date: 2026-05-05T13:39:18.939Z authors: - author: src/content/people/collin-mcnulty.md canonical_url: >- https://www.astronomer.io/blog/why-astronomer-support-is-chasing-michelin-star-service/ --- Astronomer’s Customer Reliability Engineering team was recently honored to receive [an award by Business Intelligence Group for our support](https://www.bintelligence.com/posts/2026-excellence-in-customer-service-awards-from-scripts-to-strategy-the-winners-who-proved-that-how-you-treat-people-still-wins), a step toward our audacious goal of being the best software support team in the world. We are proud to work with such an incredible customer base, many of whom use Astro for mission-critical workloads. That responsibility demands an uncommon level of service and excellence. Astronomer is the only managed service provider for Apache Airflow that provides support for Airflow itself. This presents a unique challenge for support because Dags are code, so you can do literally anything in them. This is in contrast to most software products that have only so many things the user can do with them. So, for any given ticket, we are determining whether it is an issue with the customer's Dag, an unwanted Airflow setting that we can change, an infrastructure hiccup, or a world-first bug in one of the most widely used open-source projects in the world. Also, Airflow [famously integrates with just about everything](https://airflow.apache.org/registry/providers/), so issues could be reported by Airflow but originate in other systems. We need to be able to provide a world-class experience to the customer for all of these situations. The first tool in our belt is the technical depth of our team. When I joined the team in 2021, I was interviewed by Vikram Koka, our Chief Strategy Officer at Astronomer, and he had only one make-or-break question: "What's the hardest problem related to Airflow that you've solved?" If your story wasn't gnarly enough, and didn't make him cringe, you didn't get hired. We don't ask that exact question anymore1, but we keep that spirit alive as a team that is not afraid to dig into the guts of a difficult problem. The Astronomer CRE team hires people who aren’t satisfied with the shallow answer, and who drive for real understanding. We hold ourselves to a high bar, including getting into the code of Airflow itself and our products. We don’t have a tier of support that is reliant on the documentation or a script: when you reach out to Astronomer Support, you’re getting expertise on the first contact. Of course, we don’t do it alone, so our second tool is our deep partnership with our R\&D and Documentation teams. Our team has the backing of Airflow PMC members like Jed Cunningham, Ephraim Anierobi, and Daniel Standish, who ensure that we’re up to date on the latest details of Airflow as we solve problems in the most complex enterprise Airflow environments. We also see ourselves as the eyes and ears of the organization, finding issues that real users are having, but we don’t want to keep solving the same issue over and over. We want the issue to be resolved in the product or clarified in our world-class documentation. Astronomer Support holds the product to a high bar by clearly communicating exactly what parts are causing customers to trip, so that they can be fixed. The best ticket is the one that’s never opened, so even if you never contact Astronomer Support, we’re working hard to give you an excellent experience. In 2023, we had high customer satisfaction scores and solid word of mouth recognition. But I still felt like something wasn't where it needed to be. While watching the TV show, The Bear, which depicts the great lengths restaurants go to provide a fine dining experience, I realized that the bar for software support wasn't high enough. We needed a third tool: **hospitality**. I made a kind of pilgrimage to Chicago to go to [Ever](https://www.ever-restaurant.com/) and met its then Director of Hospitality, [Amy Cordell](https://www.ever-restaurant.com/chef-curtis-and-team/)2, to understand what hospitality and service could really look like. We had been solving all of our issues with our deep technical knowledge of Airflow and cloud infrastructure, and the results were good enough to cover a gap that the hospitality industry would call “steps of service.” Astronomer CRE had always seen doing more for our customers as being more technical, being faster, taking on more scope, but we were missing whole categories of service. We started asking new questions. We asked ourselves, "how do we provide a great support experience for someone who asks a question that's clearly out of our scope and expertise?" For example, we have the humility to admit what our limits are, the courage to clearly state what we can do, and the service attitude to still give the customer everything we can to improve the issue. Another question was "how do we provide 24x7 service that doesn't feel like the customer has to re-explain the issue after every handoff?" By rethinking our setup for getting on calls with customers so that a three-touch negotiation regarding a time to meet becomes one-touch, and by writing way more internal notes. We have a lot of work to do to be in the same league as the top tier of the hospitality industry, and this third tool is very much under construction but it guides our thinking. When we think about hiring, I can teach you Airflow, but I can't teach you to truly want to help a customer and apply yourself to solve a brand-new problem for them. When we think about our proactive monitoring, how should we word our notices so that they're actionable, clear, and feel helpful? How do we use AI to empower our team, but never put the customer in the icky situation of being unsure if the person you're speaking to understands what they're saying? We think the world deserves a better class of software support. One that’s driven by deep technical knowledge, a strong feedback loop into the product, and a care for the person on the other end of the ticket. We still have a long way to go, but it's a bar I'm excited to chase for years to come. --- 1 Because of our more involved training and ramp-up program 2 She is now Chief Operating Officer