I Was Trying to Sell an Invisible Product: How I Reset My Startup Strategy

It is 2 AM in Istanbul. I have my third cup of coffee going cold on the desk. My browser has eleven tabs open, seven of which are strategy documents I wrote for myself. My Instagram account for StackMark has three followers. My waitlist has zero signups.

The validation deadline I set for myself passed eight days ago.

This is the story of what went wrong, what I learned, and how I rebuilt my solo founder strategy from scratch in a single brutal, honest conversation. If you are building a SaaS product alone, launching a designer tool, or stuck in the loop of planning without shipping, this one is for you.

The Original Plan Looked Great on Paper

When I started building StackMark, I had a plan. A real one. A 26-page Product Requirements Document with personas, competitive analysis, a SWOT analysis, a go-to-market strategy, a build timeline, revenue milestones, the works.

I had identified a genuine problem that every designer knows. I am a UI/UX designer. I had 247 design tutorials saved across YouTube, Figma community, Udemy, and various bookmarks folders. I had watched exactly 8 of them. The guilt of that number lived in the back of my head every time I opened my browser.

The idea was simple: build an AI-powered tool that helps designers actually finish what they save. Not just organize it. Finish it. A completion engine for saved tutorials, not another bookmark manager.

The positioning was sharp. The target audience of UI/UX designers was specific. The pricing was validated in theory. The tech stack of Next.js, Supabase, and the Anthropic Claude API was confirmed. I had even written a 10-day validation sprint with day-by-day execution schedules for Reddit and Instagram.

It was a beautiful solo founder strategy.

And then I executed almost none of it.

What Actually Happened: The Execution Paralysis Loop

Here is the honest timeline that every indie hacker building in public needs to hear.

Week one: I refined the strategy document instead of posting.

Week two: I rewrote the landing page copy instead of posting.

Week three: I had long AI strategy sessions to "make sure the plan was airtight" instead of posting.

By the time my February 28th validation deadline arrived, I had zero Reddit posts live, an Instagram account with three followers, and a SaaS waitlist with no signups.

I told myself the problem was the strategy. That the approach needed refining. That the messaging was not quite right. That I needed fresh eyes on the situation.

But the truth, the real truth I was avoiding, was simpler and more uncomfortable than any strategy problem.

I was scared to post. And I was using strategy as a place to hide.

This is execution paralysis. And it is one of the most common failure modes for solo founders, indie hackers, and designer-turned-entrepreneurs. The work feels real. The documents are thorough. The thinking is genuinely good. But nothing ships.

The Core Insight: You Cannot Market an Invisible Product

When I finally had an honest conversation about what was happening, one insight cut through everything else and reframed the entire situation.

I was trying to market an invisible product.

Not just in the sense that it was pre-launch. I mean truly invisible. No screenshots. No prototype. No UI design. Nothing to look at, click, or imagine using. Just words on a landing page saying "trust me, this will exist soon, and it will be great."

And I was trying to sell this invisible product to designers.

This is where the strategy had a fatal flaw that no amount of planning could fix.

Designers are the single audience on the planet that judges credibility almost entirely on visual quality. The people who look at a product and within three seconds have formed a complete opinion about whether the person who built it understands good craft. The community where the aesthetic quality of your tool is not just a feature, it is a trust signal and a credibility indicator.

I was asking designers to follow an Instagram account with no product screenshots. To sign up for a designer tool waitlist with no visuals. To believe in something they could not see.

And I was confused about why nobody was converting.

Looking back, it is obvious. But when you are deep inside a solo founder project, running on coffee and optimism at 2 AM, the obvious things have a way of hiding from you completely.

The lesson for any solo founder trying to validate a SaaS product: you cannot expect your target audience to care about something they cannot experience. Especially when your audience is visual professionals.

The Real Problem With "Building in Public" Before You Have Anything to Show

There is a version of building in public that works beautifully. You have probably seen it in the indie hacker and build-in-public community. A founder posts their MRR every week. They show the product evolving. They share design decisions with context. They document their mistakes with screenshots of the actual mistake. Their audience grows because the content is genuinely interesting and grounded in something real.

That version works because there is a product generating the content.

The version I was attempting was different. I was trying to generate build-in-public content from nothing but ideas and passion. Carousels about the problem of tutorial hoarding. Posts about learning resource statistics. Variations on the "247 saved, 8 watched" hook.

And here is the thing: that hook is strong. It is genuinely one of the best pieces of startup positioning copy I have written for this project. It is personal, specific, emotionally resonant, and immediately relatable to any designer who has ever opened their bookmarks bar.

But a hook is not a content strategy. You cannot run the same emotional beat every week for six weeks and expect a social media audience to keep caring. The hook opens the door. The product has to be standing behind it.

Without a product to show, I had one good hook and no second act. No content flywheel. No reason for designers to follow, engage, or share.

The Solo Founder Strategy Reset

Once I accepted the real problem, the solution became clear relatively quickly. And it required setting aside the original plan entirely rather than iterating on it.

Stop the validation attempt entirely. Not pause it. Stop it. The experiment had not failed. It had simply never run. You cannot declare that an approach to SaaS validation does not work when you have not actually tried it.

Then build something visual before attempting distribution again.

Specifically: design the product first. Not build it. Design it. Create high-fidelity UI screens in Figma that show exactly what StackMark looks like, how it works, and what it feels like to use. Make it beautiful enough that a designer sees a screenshot and thinks "I want that on my screen."

Once those screens exist, everything in the go-to-market strategy unlocks:

Instagram posts become product showcases instead of abstract problem statements. Reddit posts in design communities can include actual screenshots of real design work. The landing page gets genuine UI visuals instead of vague promises. The waitlist form sits beneath something that looks worth signing up for.

The content problem solves itself the moment the visual product exists. This is the real sequencing lesson for any solo founder building a designer tool without an existing audience.

Rebuilding the Product Vision and Positioning

The strategy reset also forced me to revisit something I had been slightly fuzzy on: what exactly is StackMark, at its core?

The original PRD described it as "an AI-powered bookmark manager for designers." That framing, while technically accurate, undersells the actual insight behind the product and targets the wrong emotional territory.

The real insight, the one that came from the genuine personal moment that sparked this entire project, is not about organization at all. It is about the psychology of unfinished learning and the behavioral loop that keeps people stuck.

People do not fail to finish online tutorials and courses because they cannot find them. They fail to finish them because restarting feels overwhelming.

The loop that StackMark is designed to break goes like this: you save a tutorial or buy an online course with genuine motivation. Life interrupts. You fall behind. The backlog grows. Reopening it feels heavy. Guilt appears. Avoidance begins. The guilt makes the avoidance worse. The backlog compounds. And eventually the saved tutorial joins the graveyard of good intentions you carry around with you.

That loop is not a bookmark organization problem. It is a behavioral problem rooted in decision fatigue, guilt, and the psychology of restarting. And no existing tool in the bookmark manager or read-it-later category addresses it directly.

Raindrop.io organizes your bookmarks beautifully. Readwise resurfaces your reading highlights. Pocket, before it shut down in 2026, let you read things later. But none of them answer the question that actually matters for people who hoard learning resources: how do I make restarting something I abandoned feel possible instead of heavy?

That is the product. Not "a better bookmark manager." A completion engine for saved learning content that makes restarting feel like the natural next step rather than a failure to confront.

The new positioning reflects this directly: "Stop stacking. Start finishing."

Using AI Design Tools to Unblock a Solo Founder Sprint

Once the strategy reset was locked in, I moved into a rapid design sprint using AI design tools to generate the initial product screens without spending weeks in Figma from scratch.

I wrote a detailed UI/UX design brief covering every screen, every component, the color system, typography choices, and interaction principles. Then I fed that brief into Google Stitch and asked it to design the core dashboard.

What came back genuinely surprised me.

Three of the four AI-generated screens were strong. One went full developer tool aesthetic with monospace fonts and "Command Center" navigation, which was immediately discarded. But the other three showed something important for any solo founder considering AI design tools: when you give an AI tool a specific enough brief with clear constraints, the output can be production-adjacent. Not pixel-perfect, not ready to ship directly, but close enough to real that you can start having real product conversations about it.

More interestingly, one screen Stitch generated was something I had not asked for. A "Guided Learning Journey" view. Linear steps. A current focus item with time remaining. Locked future content. A "Continue Learning Journey" call to action.

That screen was not in the brief. The AI invented it.

And it was, honestly, a more direct solution to the restart problem than the dashboard I had planned. A linear learning path with a single "continue" CTA removes decision fatigue entirely. You do not choose what to watch today. The path chooses for you.

We are not building that for the MVP. The technical complexity of automatic path generation would slow the timeline significantly, and the user experience of being told what to learn by an algorithm you just met is not ideal. But the user-created version, where a designer builds their own learning path called "UX Design 101" and manually adds tutorials to it, is a compelling v2 feature. A screen the AI hallucinated into existence became a legitimate product roadmap item in thirty seconds of founder discussion.

That is one of the more interesting things about working with AI tools at this stage of building a SaaS product: sometimes the most valuable output is the thing you did not ask for.

Execution Paralysis: The Solo Founder Pattern Nobody Talks About Honestly

I want to address something more personal here, because I think it is genuinely useful for other solo founders and indie hackers, and I have not seen many people be honest about it publicly.

I have a pattern. I suspect a lot of designers turned founders have a very similar one.

When external feedback signals are absent, I stop moving forward. And I replace forward movement with the thing that feels most like forward movement without actually being it: planning, refining, strategizing, and optimizing documents that nobody will ever read.

It feels productive. It has the texture of real work. It produces artifacts you can point to. But it is not the thing. It is a comfortable simulation of the thing.

This is a specific version of execution paralysis that hits designer-founders especially hard, because designers are professionally trained to care about quality. To not ship things that are not ready. To protect their creative reputation through the standard of their output. Those instincts are exactly right in a client work context. They become active self-sabotage in an early-stage startup context.

The mindset shift required is not abandoning quality standards. It is understanding that the MVP of your distribution strategy, your first post, your first Reddit comment, your first waitlist signup, does not need to be a masterpiece. It needs to exist and generate information.

An imperfect post that exists and gets one real comment beats a perfect post sitting in your drafts folder by an infinite margin. Because the imperfect post generates data. The perfect draft generates nothing.

What I Would Tell Another Solo Founder Stuck in This Loop

If you are reading this and recognizing your own pattern in any part of it, here is what I would tell you directly as someone who just lived through it:

The strategy session you are about to have is probably not the thing you need right now. You probably already know what to do. You are just not doing it yet.

The question worth sitting with is not "is my strategy right?" It is "what exactly am I avoiding, and why?"

For me, the answer was: I was avoiding the vulnerability of showing an unfinished thing to real people who might not care. I was avoiding the specific discomfort of posting something and hearing nothing back. I was avoiding the moment where the idea I had been carefully protecting in planning documents had to meet the world and find out if it was real.

That collision between idea and reality is uncomfortable for every founder at every stage. But it is also the only moment that actually matters. Because until you have that collision, you do not have information. You have opinions dressed up as strategy.

Go get the information. Ship the thing. Post the screenshot. Send the DM.

The rest of the plan will sort itself out once you have real user data to work with.

Current State and What Comes Next for StackMark

Here is the honest current state of StackMark as a build-in-public project:

The product is still pre-launch. The waitlist still has zero signups. The Instagram account still has three followers.

But the sequencing is finally correct.

I have a strong AI-generated design direction to refine in Figma. A detailed UI/UX brief that covers every screen decision. A strategy that puts design before distribution for the first time. A product positioning that speaks to the behavioral problem, not just the organizational one.

The next move is simple and non-negotiable: finish the Figma screens, export the screenshots, post them. No more strategy refinement. No more PRD iterations. No more sessions to make sure the plan is right before acting.

The plan is right enough. The Figma work is the only thing that unblocks distribution. Distribution is the only thing that unblocks real validation. Real validation with paying users is the only metric that actually matters at this stage of a solo founder journey.

I will document everything here on the Thnx Studio blog. The signup numbers. The conversion data. The things that work and the things that fail. No vanity metrics. No "incredible momentum" when the momentum is still three followers and a quiet waitlist.

Just the honest account of what it looks like to build a real SaaS product with a $100 total budget, a 60-hour-per-week full-time job, and the stubborn belief that the problem is worth solving.

If you have 200 tutorials saved and have watched none of them, StackMark is being built for you.

---

Frequently Asked Questions

What is StackMark?
StackMark is an AI-powered completion engine for designers who save tutorials, online courses, and learning resources they never finish. Unlike bookmark managers that focus on organization, StackMark is built around completion behavior: helping designers actually finish what they save through daily digest emails, progress tracking, and smart resurfacing.

What is the difference between StackMark and Raindrop.io?
Raindrop.io is a bookmark organization tool. StackMark is a completion tool. Raindrop helps you find what you saved. StackMark helps you actually finish it. The target problem is different: not "where did I save that?" but "why do I never go back to what I saved?"

What is the StackMark pricing?
StackMark will launch with a free tier (25 bookmarks, weekly digest) and a Pro tier at $7 per month or $60 per year (unlimited bookmarks, daily digest, full AI tagging, completion stats).

Is StackMark available now?
StackMark is currently in pre-launch validation. You can join the waitlist at stackmark.io to get early access when the beta opens.

What tech stack is StackMark built on?
StackMark is built with Next.js 14, TailwindCSS, Supabase for the database and authentication, the Anthropic Claude API for AI auto-tagging, Resend for email delivery, and Stripe for payments. It is deployed on Vercel.

What is build in public?
Building in public is a startup approach where founders share their progress, metrics, wins, and failures openly with an audience as they build. It creates accountability, builds an audience before launch, and generates genuine community around a product in development.

---

Mert is a UI/UX designer, solo founder, and the builder behind StackMark and thnx studio. He is documenting the full build journey publicly, including the failures. Follow the build at stackmark.io and on Instagram at @stackmark.io.