Tech News

I had the privilege of leading PlanGrid to $100 million in ARR before I stepped down as CEO and passed the baton to Autodesk Construction. I’ve had years to dissect the mistakes I made with my first startup.

Regardless of which industry you build in, or where you are at in your startup’s journey, there are many things that will likely fail.

This post breaks down PlanGrid’s key failure points and what I’ve learned from them. If these reflections help even one founder make one less mistake, I would consider this effort worthwhile.

Organization structure and communication failure

As first-time founders, we were too creative with our organizational structure. We had a flat management hierarchy in the early years, and we bragged that we ran our startup like “Star Trek” — you were either in engineering or operations, and everyone reported to a founder.

This was cute until it quickly stopped working. People care about titles and career paths, and if you want to retain great people, you have to care about these things too.

In Year 3, we tripled from 30 to 90 people, then doubled the team to 180 a year later. Those were the most painful years, because we went from a high-execution team to one that felt like it was stuck in molasses. We didn’t know how to hire giants, so we recruited several mediocre managers, who in turn recruited more mediocre people.

Meanwhile, communication gets a lot harder with more people, and I did a poor job communicating the direction of the company. We had a first-mover advantage in a category we created but lost our position during these years of slow execution.

Takeaways: Be creative about how you’re solving problems for your customer and not about organization structures. Hire a great HR leader as a business partner to help recruit and retain the right team and design a good communication flow. Remember that A players can recruit other A players, but B players can only recruit C players.

Internal conflict

Our trickiest inflection point was hitting Dunbar’s number — at 150 people, everything went to chaos.

Hierarchy is a factor. At 10, 20 or 30 people, everyone can report to a founder. At 150, just based on basic management ratios, the frontline team member is now separated by three to four degrees from the founders.

Not feeling like a unified team becomes dangerous when you don’t hit revenue targets or product milestones. When there is a mismatch on velocity and performance, it’s easy for those who feel like they’re performing to blame any slowdown on everyone else. There are natural tensions between sales and marketing teams, support and product, and product and engineering. Everything becomes magnified with more people simply because communication gets harder.

5 failure points between $5M and $100M in ARR by Ram Iyer originally published on TechCrunch

There’s plenty of vendors in the managed database services market — and plenty that are well-financed. Take a look at SingleStore, which last October raised $30 million to bring its database tech to new customers. There’s also EdgeDB, which landed $15 million in November ahead of the launch of its cloud database product.

Tobie Morgan Hitchcock believes that there’s room for at least one more player, though. He’s one of the two co-founders of SurrealDB, a database-as-a-service platform that provides great flexibility when it comes to querying data.

Hitchcock might be biased. But there’s third-party evidence to suggest that companies are increasingly adopting fully managed, cloud-based database services. In a recent MariaDB survey, 61% of respondents said that they were either already fully migrated or working to complete a full database migration to the cloud, motivated by a shared desire to save money and “bridge the cloud skills gap.”

SurrealDB certainly appears to be benefiting from some kind of boom. After being bootstrapped for three years (and despite being pre-revenue), SurrealDB closed a seed round recently that came in at $6 million. FirstMark led it.

“In 2015, after years of building cloud-based software-as-a-service systems with real-time APIs, complicated security permissions and multiple separate database backends, my brother Jaime and I questioned whether there might be a platform for building and scaling applications quicker while still allowing for the storage and querying of data in a structured yet flexible manner — like a database rather than an API,” Hitchcock told TechCrunch in an email interview. “We began to conceptualize and plan the SurrealDB database requirements, taking inspiration from a range of databases which we had used on previous projects.”

Prior to co-launching SurrealDB, Tobie and Jaime had tried their hand at a number of ventures including an app that let golf courses track how golfers play each hole. They also co-founded Hire Insight, a service companies can use to assess, curate and select job candidates online, akin to LinkedIn Recruiter and Workday.

With SurrealDB, Tobie says that the goal was to improve the app development process by reducing the need to build backend APIs and database layers or use a cloud data platform or single data model. (A data model is an abstract model that organizes data elements and standardizes how the elements relate to one another.) To this end, SurrealDB supports real-time queries, security permissions for multi-user access and “performant” analytical workloads, Tobie says.

Client-side apps can be built with direct connections to SurrealDB, while traditional, server-side dev setups can leverage the platform’s querying and analytics abilities. When setting up a data model, SurrealDB users can choose between simple documents, documents with embedded fields or related graph connections between records, Tobie says — depending on the nature of the data being stored.

“From the data and technical perspectives,SurrealDB offers … the ability to query one’s data in a multitude of ways,” Tobie said. “In addition, SurrealDB has the ability to handle business logic and user authentication right from within the database. Because of this, SurrealDB enables developers and organisations to simplify the backend tech stack, speed up development times and reduce the costs of complicated multi-inter-connected backend platforms.”

SurrealDB launched first as an open source database platform before transitioning into a cloud offering, although the open source package is still available and being actively developed. Current customers include “a number of” startups and “publicly-floated” enterprises, although Tobie wouldn’t name names; SurrealDB’s fully managed database service remains in beta ahead of a launch in early 2023.

“The money [from the seed round] will be used to grow the team and launch the commercial offering of [the SurrealDB cloud offering,] Tobie said. “The technology stack for SurrealDB has been purposely chosen so that SurrealDB can be quick to contribute to, easy to understand for new engineers and with as few different technologies as possible … With this in mind, SurrealDB is in a position to iterate quickly with new features and releases, whilst being able to do so with an agile and nimble team.”

For what it’s worth, FirstMark’s Matt Turck agrees. He’s an investor, of course, but he also cites the large and growing market for database-as-a-service offerings, which could be worth as much as $24.8 billion by 2025, according to Markets and Markets. Ambitiously, Turck sees SurrealDB competing not only with database vendors like MongoDB, Neo4j, Couchbase and DynamoDB but backend service providers such as Firebase, Supabase and Nhost.

“SurrealDB is an incredibly ambitious company, building a multi-model database that combines a lot of things that many thought were incompatible without major trade-offs. They’re propelled by major trends such as database abstraction, cloud and serverless,” Turck said in an emailed statement. “They’ve clearly struck a chord with the developer community — we’ve literally never seen any database open source project grow this fast. Last but not least, we loved the founding story of the company, which was started by two brothers who built the entire thing by themselves, and we see great potential in them as founders.”

SurrealDB has a three-person team at present, with the aim of increasing headcount to around 18 within the next year.

SurrealDB raises $6M for its database-as-a-service offering by Kyle Wiggers originally published on TechCrunch