I Built an Enterprise Agentic SEO Team in 12 Weeks. Now I Have to Decide Who Owns It.
Where I Am Right Now
Indexable is an enterprise agentic SEO team I built in twelve weeks — ten autonomous AI (artificial intelligence) agents plus a senior forward-deployed strategist — and I have a decision to make in the next thirty days: open-source the entire code base, or keep it closed. I'm writing this post before I've decided, not after, because I'd rather make this call with the community's wisdom than alone.
Published April 18, 2026.
Twelve weeks ago, I started building Indexable. Today, ten autonomous AI agents run inside it. One hundred twenty-eight pages live at indexableai.com. A trademark has been filed (USPTO Serial #99768585, April 2026). The product works.
I had paranoia about whether this actually holds up at enterprise level. So I ran sanity checks on every agent, multiple times. Then I extended those checks late into the night across four enterprise domains — some of the most well-known brands in the world — stress-testing the agents against real catalogs, real keyword surfaces, real competitive landscapes. They held.
And I still have zero paying customers.
Those two facts — a functioning product and no revenue — are both honest. Neither is a failure. They're the specific shape of an early-stage founder's actual life. I'm writing this post because I have a decision to make sooner rather than later, and I'd rather make it with the community's wisdom than alone.
The decision: open-source Indexable, or keep it closed.
Let me tell you why this question is harder than it looks.
What I Built
For ten years, I did one thing: enterprise SEO. I was an SEO at Uber, the first SEO hire at Uber Eats, and helped build the organic strategy for Uber Freight and other Uber domains. I was SEO Director at Zendesk, running a pure B2B SaaS program through three product lines and two acquisitions. I was Director of Technology, SEO and AI Innovation at Williams-Sonoma, leading the digital search function across a multi-brand home goods portfolio. Across those roles, my work contributed to over a billion dollars in organic revenue.
One thing I learned in all of them: the actual work of SEO is repetitive at enterprise scale, and deeply intellectual at the same time. The same diagnostic runs across every page. The same optimization patterns repeat across every category. The same competitive gaps show up across every site. But interpreting what to do about them requires judgment that takes years to build.
When large language models reached the threshold where they could reliably hold context across an enterprise site audit, the gap between "what a senior SEO does" and "what an agent could do" collapsed. I saw this coming for years. In January 2026, I decided to stop watching and start building.
Indexable's agents run across ten functions: SEO strategy, AEO (answer engine optimization — optimizing for AI answers and citations), content strategy, content engineering, technical SEO, analytics, schema and AI readability, outreach for AI authority, software automation, and ecommerce catalog optimization. Each agent operates autonomously. Each is led by a senior enterprise forward-deployed SEO strategist embedded on-site with the client team.
That last piece matters. Indexable is not a SaaS product you log into. It's a search team that deploys on Day 1 and performs like one on Day 90.
The Moral Weight
Here is the honest thing I haven't said publicly until now.
The agents I built can do the work of a senior SEO (search engine optimization) practitioner faster and more consistently than that practitioner can. Not better in every dimension. Not with the creative judgment a principal-level SEO brings. But for the routine 80 percent of the function — audits, keyword research, competitor analysis, content optimization, schema generation, data pulls, prompt research, and reporting — the agents are already ahead.
I know this because I've been that practitioner. I've been the first SEO hire at a company that didn't know what SEO was. I've been the Director overseeing a team of five. I've been the technical strategist leading a migration across a portfolio of brands. For every role I've held, my product is capable of either making the person in that seat ten times more productive, or replacing them entirely.
What the technology does vs what deployment decides
If enterprise buys my product and uses it to amplify their SEO teams, the outcome is net positive: more throughput, better decisions, humans freed from rote work to do the strategic work that actually compounds. If enterprise buys my product and uses it to cut SEO headcount from ten to three, the outcome is net negative for the seven people who used to do that job.
The weight I'm sitting with
I cannot control which decision enterprise buyers make. I can build the product. I can influence how it's positioned. But the deployment choice is theirs.
This is the weight I'm sitting with. It's not theoretical. I'm pitching CMOs and CEOs right now. Some of them are asking me the exact question I'm asking myself: "If we deploy this, what happens to our team?" I've been giving them the honest answer — "that depends on what you decide this technology is for." But the honesty doesn't solve the problem. It just names it.
Why Open Source Is on the Table
The original plan for Indexable was straightforward: build a closed, venture-backed platform. Raise a seed round. Sell to enterprise. Scale.
Here's what changed my thinking.
A few weeks ago, OpenAI acqui-hired Peter Steinberger and his open-source AI agent OpenClaw — a project that started as a weekend experiment, grew into an open-source tool used by thousands of developers, and became valuable enough that OpenAI paid serious money to bring both the code and the creator inside their ecosystem. Peter built something the community adopted, and the adoption itself was what made it valuable.
That story reframed a question I hadn't been asking: what if the most durable way to lead a new category is to give away the core and capture value from the expertise around it?
The case for open-sourcing Indexable
| Argument | Why it matters |
|---|---|
| AI moves weekly | Proprietary code calcifies. An open codebase gets community contributions, and the code that updates fastest wins. |
| Category creation compounds | If the agents are open, the community extends them and writes the textbook. Closed keeps the category small. |
| Enterprise trust is shifting | Sophisticated buyers prefer open-core over black-box AI vendors. Mind-share is winning the market. |
| Talent pulls toward open | Engineers I'd want to hire prefer public codebases that build their reputation. |
I've been writing publicly about Agentic SEO — the category Indexable is building into — and about the agentic web that sits underneath it. In a market where CMOs (chief marketing officers) are being asked for specific answers on AI search visibility, buyers want to see inside the system that produces those answers.
The case against
| Risk | What breaks |
|---|---|
| Revenue dilution | Open-source contributors may use the free layer and never pay for premium. Business model rebuild required. |
| Competitor weaponization | Direct competitors could clone the code and differentiate on sales motion alone. |
| Messy unit economics | Closed SaaS (software as a service) has clean unit economics. Open-core services are harder to model, harder to pitch to investors. |
| Investor narrative | Some VCs invest in moats. Open source looks, at first glance, like no moat. |
I don't know which path is correct. I know both paths have won for other founders in adjacent categories. I know both paths have failed for other founders in adjacent categories. My specific situation — pre-revenue, category-defining, early-stage — has arguments both ways.
How You Can Help Me Decide
I'm not asking for advice. I'm asking for wisdom. Here is how you can help, in three specific steps.
Step 1: If you're an enterprise buyer of marketing or AI infrastructure, tell me: would you trust a closed platform or an open-core vendor more for something as strategic as your AI search visibility? You should give me the specific reason, not the philosophical one.
Step 2: If you're a founder who has faced this decision, tell me what you decided. What changed your mind? What do you wish someone had told you before you decided? You can share privately or publicly.
Step 3: If you're an AI or SEO practitioner, tell me if I'm wrong about the replacement question. Is the right answer "the field adapts, as it always has," or is there a real and specific harm I should be weighing differently? Then implement whatever correction makes the most sense in your experience.
If you're someone I've never met who's thought about any of this: I want to hear from you most of all.
You can reach me three ways. DM me on LinkedIn. Email me at [email protected]. Or just comment on this post where you found it. Start by sharing one specific reaction — I'll reply to every message.
Conviction Despite Uncertainty
This isn't a cry for help.
Today, right now, while this post is live on the internet, I'm sending outreach emails to CEOs at companies I want to work with. I'm refining the product. I'm prepping for meetings with people who might become the first Indexable customer. I'm doing the work that makes this decision worth having.
What I'm not doing is waiting for the decision to become obvious. It won't. Both paths are live. Both paths demand conviction. The right question isn't "which is safer" — it's "which one do I believe in enough to defend for the next five years."
I'm going to make this call sooner rather than later. When I do, I'll write a second post about what I decided and why.
Until then — if you've read this far, thank you. I'd value whatever you can share.
Frequently Asked Questions
Why consider open source if you're pre-revenue?
Because category creation compounds under open source, and AI search is a category-defining window that closes in 12–18 months. Pre-revenue open-sourcing is counterintuitive but potentially strategic — the OpenAI/OpenClaw acquisition pattern (Steinberger, 2026) suggests the community-adoption path can produce outcomes a closed startup can't.
Does open-sourcing mean free?
No. Open-core means the code is open; paid services, integrations, and forward-deployed strategy remain commercial. I would keep the execution layer (the work I do alongside the agents) paid. The agents themselves are what the community would extend.
What would the acquisition path look like?
The closest precedent is OpenAI's acqui-hire of Peter Steinberger (founder of OpenClaw, February 2026). An open-source project with enterprise traction becomes acquisition-interesting for the AI labs whose ecosystems it plugs into. That's a possibility, not a plan.
Who is the right audience for this post?
Founders who've faced a similar open-vs-closed decision. Enterprise buyers of AI infrastructure. AI and SEO practitioners. Anyone who has opinions on how categories form around open-source infrastructure.
When will you decide?
Within thirty days of publishing this post. I will publish a second post with the decision and the reasoning.
Have a view on this?
I'm reading every DM, email, and comment. The decision gets better with more voices.