Loading page…
A Photoshop plugin that turns an hour of manual mockup work into two minutes, sold to print-on-demand sellers for $15/month
"I do not brainstorm content ideas. I harvest customer pain." A 370-view YouTube video brought him three customers — $45/mo, forever, from one recording nobody watched.
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
Vikash started as a Photoshop freelancer on Upwork and freelancer.com. One day a client wanted part of a Photoshop workflow automated. He didn't know how. Instead of skipping the job he told the client to give him time, went to Stack Overflow, learned enough JavaScript in a single day, and by that night had hacked together a Photoshop script that did the job. It became an internal tool for his own freelance gigs. The commercial moment came from a different job posting. Clients were hiring virtual assistants to create mockups for their print-on-demand stores. He applied for one that asked for 1,800 mockups — work the client expected to take three or four days. His script finished all 1,800 in under thirty minutes. The client was stunned and asked how he'd done it. Vikash didn't explain; he said give me $300 and I'll give you this. The client wired the $300 without a second thought. That was the moment he knew. The first version was a hacky script sold as a one-time fee. He realised that wasn't defensible, locked himself in his room for two straight months, and wrapped it in what he describes as a very bad UI. He didn't know how to do licence validation or user management, so for a couple of years he could only sell it as a lifetime deal. Once he had enough cash he went back to Upwork, hired a developer, and had the product polished into what it is today. It now does $12,000-$13,000 a month, consistently, shown from his Gumroad dashboard.
The first customers were already paying for the work by hand. Vikash was taking Photoshop jobs on Upwork and freelancer.com when he noticed clients hiring virtual assistants to produce mockups for their print-on-demand stores.
The quick-reference card — acquisition channels, replication playbook, risks & traps, and our verdict — is members-only.
Join the Unicorn Club — $5/moCancel anytime, in one clickPaste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself. The exact words are quoted below, with a timestamp that jumps to the second each figure is said.
“This is Vikash, a guy from India who built a simple piece of software that makes $12,000 per month.”
0:02 · hear it in the videoData credibility: Founder interview on Starter Story. Vikash showed his Gumroad dashboard on camera covering the prior year at a consistent $12,000-$13,000/month — that is the strongest evidence here, but it is a self-shown dashboard, not third-party data, so we grade it founder-reported. The per-video returns ($45/mo from a 370-view video, $213 over six months from a 12,000-view video) and the tool costs are stated verbally and unverified. The $15/month price point is only mentioned in passing while he does the maths on one video; the full pricing tiers are never shown, so the implied subscriber count (~800) is our arithmetic, not his. His name is rendered variously as Vikash/Vikos/Nikkosh in the auto-generated transcript.
The first dollar was $300, and it was not a subscription. It came from the Upwork client who had just watched a four-day job finish in under thirty minutes and paid for the script itself, with no negotiation.
The flywheel is explicit, and he names its fuel: "I do not brainstorm content ideas. I harvest customer pain."
He had one piece of evidence to hold onto, and it was the $300 wire.
Bulk Mockup is a Photoshop plugin. The customer sets up three folders: PSD mockup templates, design files, and an output folder. They point the plugin at the template folder and the design folder, hit start generating, and it opens each template, drops each design onto it, and exports the result. In the demo he runs five templates against ten designs — fifty mockups — in about a minute. Done by hand, placing each design into each smart object and exporting, he estimates thirty to forty minutes. On the original 1,800-mockup job the ratio was three to four days of human work versus under thirty minutes. The customer is a print-on-demand seller — Etsy, Redbubble, Printify shops — who needs product photos of every design on every garment colour and style before they can list. Mockup generation is pure repetitive labour that sits directly between them and having a sellable listing, which is why they'll pay a monthly fee for it. Our read: this is the cleanest possible shape for a micro-SaaS. It does one thing, the before-and-after is visible in a screen recording in under sixty seconds, and the value is measured in hours saved per week by someone whose income depends on listing volume. Vikash's own test — does your tool make people money, save them money, or save them time — is answered on all three.
Customers find him on YouTube, buy the plugin, run it inside their own Photoshop with their own mockup templates, and pay monthly. Sales run through Gumroad, whose dashboard he shows on camera covering the last year at a consistent $12,000-$13,000 a month. Pricing is only stated indirectly: when he works out the return on a single YouTube video he says three customers at "$15 a month" — so the entry price point is $15/month. At $12,000/month that implies roughly 800 subscriptions at that tier, or fewer at higher tiers; he never breaks out the plan structure. The monetisation history is instructive. Version one was a one-time fee, which he abandoned because it wasn't defensible. Then, because he genuinely couldn't build licence validation or user management, he sold lifetime deals for a couple of years — trading long-term revenue for the capital he needed to hire a developer on Upwork and build the real product. Only then did it become a recurring-revenue business. Our read: the lifetime-deal phase is usually treated as a mistake in indie-hacker circles. Here it functioned as bootstrap financing, and it worked because the product had already proven demand at $300 a pop.
This is the part worth stealing. Vikash calls it a content flywheel, and the fuel is customer pain, not keyword volume. He makes YouTube videos that solve one specific long-tail problem for one specific customer. The videos do not go viral and he does not want them to. His two worked examples: a video on splitting a design manually and automatically in Photoshop has around 370 views over four months and produced three customers. At $15/month that's $45/month of recurring revenue from one recording with 370 views. A second video hit 12,000 views over six months and produced $213 in six months. The low-view video is worth more per view by an order of magnitude, because it was aimed at a problem rather than at an audience. He harvests the pain from four touchpoints. One: he reads communities where his customers hang out, silently, listing pains that repeat day after day — he never sells there. Two: onboarding emails that ask every new signup whether they want a custom tutorial for their situation. Three: customer support, which he explicitly runs as an education channel — when someone has a problem he asks them to send their file and records a custom Loom for them, and sometimes gets on a call and walks their whole workflow. Over three years that's produced more than 1,500 recorded conversations, which he calls a gold mine of edge cases that never show up in keyword research. Four: YouTube comments — he finds videos in his niche with low views but high comment counts, because that signals unresolved pain, lists every objection, and makes better content answering them. The distribution step matters too: he optimises each video for Google, not just YouTube, because Google ranks video results first for how-to queries. 22% of his channel's views come from Google search.
The product is a Photoshop plugin, so there is effectively no infrastructure to pay for — the compute happens on the customer's own machine, running against the customer's own Photoshop licence and their own templates. That is the structural reason the margins are what they are. His stated running costs: Komodo Decks (a Loom alternative for recording customer tutorials) bought as a $19 AppSumo lifetime deal. A helpdesk for support ticket management at about $15/month. A reviews widget for the website. Zoom at about $20/month for support calls. Adobe at $40/month. A screen recorder for YouTube tutorials. Notion at $12/month holding the content system. Cal.com at $0 for scheduling. That totals roughly $90-$110/month in tooling against $12,000-$13,000/month in revenue, plus Gumroad's transaction fees. There is no server bill, no AI inference cost, and no paid acquisition spend at all. He mentions hiring a developer from Upwork once, historically, to polish the product. Our read: near-total margin is what makes the low-view YouTube strategy rational. When a video that costs an afternoon produces $45/month of recurring revenue at ~95% margin, and the video keeps producing indefinitely, you don't need the video to perform.
Copy the discovery method first. Vikash found this business by doing the manual work himself on Upwork and noticing that clients were hiring humans to do something a script could do. Freelance marketplaces are a live feed of tasks people are already paying money to have done — which is a far better validation signal than a survey. Copy the content flywheel exactly as described: harvest pain from communities, onboarding emails, support calls and high-comment low-view videos; make one video per specific pain; optimise the title, description and first thirty seconds of transcript for the search term; publish and let it sit. Do not chase views. His three-step process is collect pain, create content, distribute for search — and the third step is the one most people skip. Copy the support-as-content loop. Recording a custom Loom for a confused customer costs fifteen minutes and produces both a retained customer and a piece of content. Three years of that is 1,500 recordings' worth of raw material. What to change: don't do the two-year lifetime-deal detour. Licence validation and user management are a weekend with modern tooling, so start recurring on day one. And pick a different host platform — Photoshop is one option, but the same shape exists as a Figma plugin, a Canva app, a Blender add-on, an Excel add-in, or a Shopify app. The pattern is: find a professional tool where users do a repetitive batch job by hand, and automate exactly that batch job.
Platform dependency is the obvious one. The product lives inside Photoshop. Adobe controls the plugin APIs, has deprecated extension architectures before, and is shipping its own generative and batch features. A native bulk-mockup feature in Photoshop, or a Firefly workflow that does the same job, would remove the reason to buy. AI image generation is the second, larger threat. The job the plugin does — put this design onto this garment photo — is exactly the kind of task that image models are getting good at, and several print-on-demand platforms now generate mockups automatically at listing time. The moat here is not the algorithm, it's that sellers already own libraries of specific PSD templates and want their exact templates used. Third: single-channel dependency. Effectively all acquisition is YouTube, and 22% of that is Google surfacing YouTube results. Two algorithm changes at one company could take out the whole funnel. There's no email list strategy, paid channel, or affiliate programme described. Finally, the founder himself named the personal risk on camera: he ground fourteen to sixteen hours a day for years and had spine surgery last year. His own first piece of advice to anyone copying him was don't ignore your health.
The four founder playbooks — first customers, first dollar, the flywheel, the silent stretch — are in the Unicorn Club, on this idea and on all 1,403 others. The story, the numbers, the sources and every later section stay open to everyone.
Join the Unicorn Club — $5/moCancel anytime, in one click