How Endra Closed $1.5M ARR in One Month With Enterprise AI

Niklas Lindgren spent $150K on a website before Endra had a product — against every early-stage playbook. Six months later, he closed $1.5M in enterprise ARR in a single month. Here's exactly how he did it.

Key Facts

• January 2026 ARR: $1.5M closed in a single month, all enterprise (Niklas Lindgren)
• Series A raise: $50M led by a16z (June 2026), following $4M pre-seed and $20M seed (Niklas Lindgren)
• Early brand investment: ~$150K spent on website, visuals, and storytelling before a product existed (Niklas Lindgren)
• Average contract value: ~$300K ACV at time of first enterprise closes (Niklas Lindgren)
• Market size: MEP consultancy market ~$150B; MEP equipment market ~$1 trillion (Niklas Lindgren)

The $1.5M Month: How Enterprise FOMO Drove Endra's PMF Moment

Endra closed $1.5M ARR in January 2026 — all enterprise — by starting commercial conversations in late 2025, running fast pilots in early January, and converting them to signed contracts within weeks. Competitor fear among large MEP firms accelerated every deal.

Endra's product market fit moment was unusually clean: pilots started in early January 2026 and converted to signed contracts by month end. The initial ACV was around $300K, targeting large MEP engineering consultancies in the Nordics and UK. Four to five enterprise customers in a single month is rare for any early-stage company, but Niklas had a clear explanation for why it happened so fast.

The dynamic was competitive panic. Once one major consultancy began a successful pilot, its rivals needed to respond. As Niklas put it, if a competitor 'can deliver their work much faster, you can deliver your work much better at a cheaper price point' — the stakes were existential. Enterprise buyers who normally move slowly had a concrete reason to move fast: their number one competitor was already doing it.

This kind of sales motion — where your customer's competitors are your best channel — is hard to engineer but powerful when it exists. Endra's early GTM was mostly outbound, then flipped almost entirely to word of mouth. The competitive structure of the MEP industry, where a handful of global consultancies compete for the same contracts, made that flywheel spin quickly.

"My number one competitor is doing work with these guys and it actually seems to work, and they're doing successful pilots. They wanted to speak with us." — Niklas Lindgren

"We closed a million and a half in January alone. And that was all enterprise revenue, which was a very strong signal that the global industry would want to do this." — Niklas Lindgren

What Endra Actually Does: AI for the $150B MEP Engineering Market

Endra is an agentic design platform for MEP (Mechanical, Electrical, Plumbing) engineers. It replaces a fragmented five-tool stack built around Revit — software from 1997 — with a single AI-powered platform that can auto-generate building system designs in minutes instead of hundreds of hours.

MEP engineers design everything inside a building that isn't the structure or architecture — HVAC, electrical systems, plumbing. Every fixture, receptacle, and lighting switch in a hospital is manually placed in a 3D model before installation. According to Niklas, a project that previously took 400 hours can now be completed in Endra in under ten minutes.

The core workflow problem was fragmentation. Engineers worked primarily in Revit — built in 1997, acquired by Autodesk in 2002 — plus four or five siloed tools that didn't communicate with each other. Moving data between systems was largely manual copy-paste work. Endra bundles this into one platform and runs agents to generate designs based on physics and simulation.

Critically, Endra didn't try to replace Revit outright. It integrated natively with legacy tools, the same way Harvey and Legora integrated with Word for legal workflows. The output from Endra translates directly back into existing platforms, removing the switching cost barrier that kills most enterprise software pitches.

"When engineers are starting to say, holy shit — I know because they gave us their models that they previously completed — once you saw, once I ingest this into Endra, it comes to eighty, ninety percent of whatever I did before and I spent four hundred hours doing this project. Endra created this for me in under ten minutes." — Niklas Lindgren

"As Legora and Harvey integrate natively into Word, we did the exact same thing. So the user who works in Endra, whatever work is being done there is perfectly possible to just translate into these older platforms." — Niklas Lindgren

• Revit (1997) is the industry standard — widely disliked for its legacy architecture.
• Engineers manually place tens of thousands of units per project across disconnected tools.
• Endra generates 80-90% of a design automatically, which is a complete first deliverable in an iterative process.
• Native integration with legacy tools removed adoption friction.

Why Spending $150K on Brand Before a Product Was the Right Call

Niklas invested ~$150K in brand, website, and storytelling before Endra had a product — funding a visual identity, product UI concepts, and narrative assets that let engineers understand and believe in what was being built. That credibility opened enterprise conversations months before launch.

This decision went against almost every early-stage playbook. Niklas acknowledged it felt strange at the time: 'It sort of went against all early stage playbooks of how you should do things.' But the bet was that in a market where nobody knew Endra existed, the brand needed to do the work that a product demo couldn't yet do.

Stockholm-based agency EW built the website, motion visuals, and early product UI concepts. The result wasn't just aesthetics — it was a structured story. Niklas described it simply: 'A good story always, to some extent, has a start, a middle part and a finish.' The website explained current state (how engineers work today), the problem (why that's broken), and the vision (what the future looks like). That narrative gave engineers and enterprise buyers something to believe in before a single line of working product existed.

The downstream effect was a waitlist and word of mouth that seeded the eventual GTM. Engineers who found the site shared it. LinkedIn followers grew. Industry voices started tagging Endra in posts. The brand investment meant that when Endra started outbound sales conversations in late 2025, the story was already in the market.

"We got an identity for those $150k and that was very, very important for us and it's paid off multiple times after." — Niklas Lindgren

"People started believing in the work that we did. They were very excited about the work that we did and that's also how we got into very, very intimate discussions with the ICP early on." — Niklas Lindgren

How Reddit DMs Became a $300K Enterprise Sales Channel

Endra's earliest users came from direct messages in MEP engineering Reddit threads. Niklas used 'shadow marketing' — commenting and messaging in relevant communities — to build a waitlist and identify power users, who then became the conduit into enterprise decision-makers.

Without an existing network in MEP engineering, the Endra team went where engineers actually talked: Reddit. They posted in MEP engineering threads, commented on discussions, and sent direct messages to practitioners asking questions about workflows and validating early concepts. Some conversations stayed on Reddit; others moved to calls.

Niklas was direct about the tactic: 'We were commenting on threads on Reddit and doing a bit of sort of shadow marketing. I know you shouldn't do that, by the way, on Reddit but we did early on in some threads. That was how we kickstarted it.' Reddit surfaced genuine practitioners — the actual users — even if the enterprise decision-makers (heads of engineering at large consultancies) weren't active there.

The Reddit strategy worked as a two-step: find engaged individual engineers, show them an early prototype, let their excitement create internal champions. Those champions then pulled Endra into enterprise conversations from the inside. The word-of-mouth loop that eventually drove $1.5M in a month started with individual DMs in niche forums.

"Most of our conversations early on were on Reddit, with DMs on Reddit, in MEP engineering threads." — Niklas Lindgren

"Talk to your customers like hell and try to get every piece of information you can from them. I did it on Reddit, I did it on cold calling, emailing, going through contacts, whatever." — Niklas Lindgren

Hiring Principles: How Endra Screened the First 50 Employees

After reaching 10-11 people, Endra formalized a hiring principles document with 10 criteria — including 'talent magnet' and 'energy in, energy out' — that every candidate must pass. Founders did all early LinkedIn outbound themselves, screening for life trajectory rather than résumé credentials.

Niklas built Endra's hiring framework from a clear belief: the first 50 employees define the culture and capability ceiling of the company. The principles document, created when the team hit 10-11 people, gave every hiring decision a consistent filter regardless of which founder was running the interview.

Two principles stand out. 'Talent magnet' asks whether the candidate has a network of other exceptional people who want to work with them — a proxy for reputation and collaborative track record. 'Energy in, energy out' screens for whether someone adds net positive energy to the room. All 10 principles are assessed, though only 3-4 are specifically screened for in interviews; the rest emerge through the process.

The interview technique Niklas relies on is the life story: 'Can you tell me your story, from early on — who were you as a child and who were you as a teenager and how are you getting into this position today?' This open-ended format lets candidates self-select the details they reveal, and gives interviewers a long timeline to find evidence of exceptional behavior, failure, and growth. For sourcing, founders personally ran LinkedIn outreach — no recruiters, no agencies — until the team was large enough to leverage employee networks.

"We want you to at least have access to other people. Because that signals that people want to work with you and that you're a good person." — Niklas Lindgren

"Everyone needs to pass those hiring principles. Otherwise, they don't get hired." — Niklas Lindgren

• 10 hiring principles; 3-4 actively screened per interview.
• 'Talent magnet': does this person attract other great people?
• 'Energy in, energy out': net positive presence in the room.
• Life story interview reveals trajectory, depth, and self-awareness.
• Founders personally ran LinkedIn outbound for the first 50 hires.

Endra vs. Legacy MEP Workflow