Zero to $10M ARR in 2 Years: Aragon AI Founder Wesley Tian

Wesley Tian was fired from his job, down to his last few thousand dollars, and had been rejected by 30 VCs when Aragon finally launched in February 2023. Two years later, he's running a $10M ARR, profitable AI headshots business with just 10 people.

Key Facts

• Time to $10M ARR: 2 years (Feb 2023 – Feb 2025) (Wesley Tian)
• Team size at $10M ARR: 10 employees (Wesley Tian)
• Revenue growth: $3K → $25K → $50K → $180K/month in under 6 months (Wesley Tian)
• Single affiliate post contribution: Over 50% of early revenue from one blog post ranking first for 'AI headshots' (Wesley Tian)
• VC rejections before launch: 30 VCs said no before Aragon generated its first dollar (Wesley Tian)

How Wesley Found the AI Headshot Opportunity in 2022

Wesley and his co-founder were testing AI image models — Stable Diffusion and Google's DreamBooth paper — and built a multi-tool app store to find demand signals via waitlists. Over 50% of waitlist signups chose professional headshots, making the focus obvious.

Before Aragon existed, Wesley and co-founder Akhil were running a classic idea maze — exploring HR tech, agriculture software, and AI image tools on weekends while working full-time. The turning point came when Akhil spotted two technical breakthroughs on Twitter: Stable Diffusion, the first strong open-source image model, and Google's DreamBooth paper, which showed how to teach an AI model what a specific face looks like.

Rather than building one product, they created a multi-tool website — a sidebar with different AI image packs including anime photos, logo generation, product photography, and professional headshots. Most were waitlists. One was live. The data was unambiguous: professional headshots dominated demand. Wesley recalls that more than 50% of all pack selections were the headshot option, even when users had four free choices. That one data point drove their entire pivot.

They incorporated in October 2022, launched the focused headshot product in February 2023, and went all-in on a category that still looked like a toy to most observers.

"Over 50% of people only chose a professional headshot pack — which told us this pack was worth four times as more than other packs." — Wesley Tian

"I think the best time to build is when there's a technical innovation — a new AI model, or a regulatory change. This was the former." — Wesley Tian

Why People Paid for a Bad Product — and What That Means for PMF

Even with terrible output quality — distorted faces, extra hands, unusable photos — users still paid and shared Aragon because it was nearly 10x cheaper than a traditional photographer and removed enormous friction. Value proposition beat product quality at launch.

The early Aragon product was, by Wesley's own admission, atrocious. Out of 250 generated photos, most users couldn't find a single usable one. There were photos with three hands, two heads, misaligned irises. Yet people still paid $30, shared the product with friends, and stuck around.

The reason was simple math: traditional headshots in the US cost around $250. Aragon charged $30. That's nearly a 10x cost saving. Beyond price, the friction of hiring a photographer — finding one on Yelp, booking a session, wearing a suit, traveling — was replaced by uploading selfies from your phone. Wesley managed the quality gap through an aggressive refund policy and hands-on customer support, sometimes spending 30–60 minutes per customer helping them manually crop and improve photos.

This early traction taught Wesley something critical: if the value proposition is strong enough, imperfect execution still creates a market. The product improved rapidly as new model versions released, but the customer appetite validated the direction from day one.

"For a long time, most users would actually not get a single photo out of 250 that they could use. It was that bad." — Wesley Tian

"Even though the landing page was pretty bad and the photos were bad, people were still sharing with other people and still willing to pay. And the reason for that is because it's a 10x cost savings." — Wesley Tian

• Traditional headshot cost: ~$250. Aragon at launch: ~$30. Nearly 10x cheaper.
• Refund rate dropped from 10%+ early on to 1–2% as product quality improved.
• Wesley personally handled 10–20 hours of customer support daily during early growth.
• 90% of current users download at least one photo they like, up from near zero at launch.

The Affiliate Marketing Move That Drove 50% of Early Revenue

Wesley set up a simple affiliate program using Rewardful (20% revenue share), and one blog organically ranked Aragon first in a 'top AI headshot generators' roundup that itself ranked first on Google for 'AI headshots' — a single post that drove over half of early sales.

Wesley's growth stack was minimal: a Twitter post, some LinkedIn content, guerrilla Reddit and Blind posts, and an affiliate program. The affiliate program was set up on Rewardful with a 20% commission on every sale. He didn't manually recruit affiliates — they came organically, attracted by the product's novelty and the commission incentive.

One blog site, which Wesley calls 'Geek Flare,' created a roundup article titled something like 'Top AI Headshot Generators' and ranked Aragon first. That article itself ranked first on Google for the keyword 'AI headshots.' The compounding effect was immediate and enormous. Revenue jumped from $2K in March 2023 to $25K in April — a 10x spike — attributed largely to that single post. Aragon has since paid that one blog over $100,000 in affiliate commissions.

The affiliate backlinks also boosted Aragon's domain authority, which improved organic rankings further. Wesley notes that because they launched early, they built authority before competitors existed — making it structurally difficult for later entrants to outrank them.

"That one blog post was generating over 50% of our revenue at one point — because it ranked first for AI headshots. We've paid out over $100,000 to that blog site through affiliate commissions." — Wesley Tian

"We didn't really source backlinks ourselves. We didn't even understand the concept of backlinks. It was just a landing page that people liked and converted on. Google looks at time on site, conversions, bounce rate." — Wesley Tian

The Revenue Growth Curve: From $3K to $10M ARR

Aragon went from $3K in February 2023 to $180K/month by July 2023, plateaued around $2M ARR through late 2023, spiked to $7–8M ARR run rate in early 2024, and hit $10M ARR by Q1 2025 — driven by SEO, Google Ads, word of mouth, and seasonality.

Wesley shared exact revenue figures rarely disclosed by founders at this stage. February 2023: $3K. March: $2K (post-launch dip, no consistent traffic source yet). April: $25K (affiliate post goes live). May: $50K. July: $180K. The product was a one-time purchase for most of 2023, which meant revenue was entirely dependent on new user acquisition — no recurring cushion.

Traffic sources today break down roughly into thirds: SEO (~33%), Google Ads (~33%), and word of mouth (~33%). Seasonality plays a major role — Q1 is peak season as people update LinkedIn profiles and search for jobs in the new year. Wesley estimates non-Q1 quarters run at roughly 70–80% of peak rates.

The business runs on 10 people and is 'pretty profitable' — with Wesley confirming profits are over seven figures annually. The lean structure is intentional: no office, no swag, no sales team. Every dollar went back into product and traffic.

"February was $3K, March was $2K, April was $25K, May was $50K, July was $180K. It basically just doubled every month." — Wesley Tian

"It was non-recurring. Everyone I talked to was like, oh, it's cool, it's a nice project, but it's not recurring. How are you going to scale? Now we're at $10 mil. Who's talking now? Nobody." — Wesley Tian

What Product-Market Fit Actually Felt Like for Aragon

PMF arrived when Aragon's servers broke under unexpected demand during a NEA accelerator retreat in Oregon. Co-founder Akhil — an early Brex engineer — said it felt exactly like Brex's hypergrowth phase: pure market pull, reacting instead of planning.

Wesley describes two signals that confirmed PMF. The first was visceral: their backend collapsed under callback overload when revenue jumped 10x in a month. Co-founder Akhil spent 36 hours straight in a cabin in Oregon rewriting their entire infrastructure while Wesley brought him food — all while NEA partners were asking where Akhil was.

The second signal was conceptual. Akhil had been an early engineer at Brex during its run from a few million to $100M ARR in roughly 18 months. For the first time at Aragon, he felt the same sensation — market pull so strong you stop building proactively and just react. Feature requests poured in. Users kept coming without being pushed. Growth felt like it was happening to them, not because of them.

Wesley is clear that before this moment, they doubted everything. The one-time payment model made every month feel like starting over. They worried the affiliate blog could pull their listing. But once the servers broke, the doubt lifted.

"For the first time working on Aragon, my co-founder started to feel the same feeling he had at Brex. It was like a pull from the market — you can't plan ahead. You're just reacting." — Wesley Tian

"It's like the market is pulling you. The sheer demand of users — you have to fix all these problems. And people just don't stop coming. It just keeps growing." — Wesley Tian

Founder Lessons: What Wesley Would Tell Every Early-Stage Founder

Focus only on what moves the business: building something people want to pay for, driving traffic, and optimizing conversion. Ignore logos, swag, and offices. And protect your health and relationships — burnout without a support system is an existential risk.

Wesley's framework is ruthlessly simple: two variables determine revenue for a non-subscription product — traffic and conversion rate. Everything else is noise. He ran Aragon for years without an office, without swag, without a sales team. Even profitable and scaling, he ate PB&Js, took the bus without paying, and kept burn at $2,500/month in San Francisco.

On the product side, his advice is to launch with a waitlist or a small charge to validate real demand before building. He used The Mom Test framework and built landing pages before features. On conversion, he cites social proof as the highest-leverage change — adding 'used by people from Microsoft' or 'used by 100,000 people' materially improved conversion even before those numbers were fully accurate.

His most personal advice is about burnout. After a year of skipping workouts, losing his friend network, and tying his emotional state to revenue charts, Wesley burned out badly — losing 10 pounds of muscle and gaining 10 pounds of fat. His warning: when revenue dips (even seasonally), if your company is your only source of stability, you have nothing to hold onto.

"The only thing that matters, if you really care about building a business, is building something that people want, building something that people want to pay for, getting distribution, getting traffic, conversion rate optimization, building the best product out there. Everything else — whatever. Name of the company, logo, swag, office — we don't have an office." — Wesley Tian

"When my company started plateauing and revenue started dipping, I didn't know it was seasonality. My emotions were dipping with it. I had nothing else to keep me afloat." — Wesley Tian

• Traffic × conversion rate = revenue. Master these two before anything else.
• Social proof on landing pages — logos, user counts — drives conversion even early.
• A co-founder creates accountability; working alone is structurally harder.
• Burnout compounds fast when work is your only identity or support system.
• Ignore company name, logo, office, and swag until you have strong PMF.