Elephas: 14 Questions Answered From the Starter Story Interview, With Timestamps
Elephas is a Mac app that answers questions from your own files and notes, and its marketing co-founder grew it in two stages: Reddit first, for feedback and early revenue, then search content once Reddit stopped scaling. In the Starter Story interview, co-founder Ayush reports going from zero to $150,000 in annual recurring revenue over three years, with Reddit taking it to $3,000 in MRR and search content taking it to about $12,000 a month.
We transcribed the interview with transcribe.so and asked it the same questions we ask every episode. Each answer below is written in our own words and points to who said it and the exact second, so you can check it in the video. All figures are self-reported by the founder and we have not audited them. We are not affiliated with Starter Story, and we sell a transcription service. The interview does not say who the ideal customer is or how that group was found, so that question is left out.
| Who is speaking | Role |
|---|---|
| Ayush, co-founder of Elephas | Answers the questions |
| Starter Story hosts | One asks the questions; a second joins him for the closing reflections |
What does Elephas do and how does it make money?
It is an assistant for the Mac that indexes your local files, such as PDFs and notes (Founder, 01:37), so you can build knowledge bases, chat with them and draft content and reports (Founder, 01:46). It sells monthly and annual subscriptions plus lifetime deals, through its own site, an app subscription bundle and the Mac and iPhone app stores (Founder, 02:20).
What numbers does the founder share?
Zero to $150,000 in annual recurring revenue over the last three years (Founder, 01:18), now about $12,000 a month (Founder, 05:03). On his dashboards: about 110k in license sales this year and about one and a half thousand from the Mac app store (Founder, 02:20), and about 180,000 website visits a year (Founder, 02:41). The host's intro puts it at over $150,000 a year (Host, 00:02).
Who is the founder?
He had just left an 11-year corporate job and was consulting part time for early-stage startups when his business partner, who started Elephas in 2022, hired him as a consultant (Founder, 02:55). After five or six months of good results he joined as co-founder and the marketing half of a two-person team; his partner writes the code (Founder, 03:08).
Where did the idea come from?
The interview does not cover the original idea, which predates his arrival. What it does cover is how the main feature took shape: a knowledge-base feature they demoed on Reddit drew so much feedback and so many requests that it became the core of the product (Founder, 07:33).
How long did it take to build, and with what?
It is a native Mac app written in Swift (Founder, 17:31). The interview does not say how long the first version took.
Why did Elephas grow on Reddit first and then switch to SEO, and what did each phase bring in?
Reddit came first because niche communities give fast feedback and early buyers (Founder, 03:48). That phase took them from zero to $3,000 in MRR in about six months, before returns faded (Founder, 04:08). Search started by accident: help articles written for early customers began ranking, and over 2023 search alone brought in about $70,000 (Founder, 04:36), taking them from about 3K a month to about $12,000 (Founder, 05:03).
What single thing moved the numbers most?
He names no single winner, but two examples stand out. On Reddit, a post that wrapped a video demo in a problem-and-solution story, with a free trial and a request for feedback (Founder, 06:39). In search, a simple help article that ranked first for six to eight months (Founder, 11:52) and later a roundup of the best chatbot apps for Mac that ranked first and brought high-intent traffic (Founder, 12:46).
"Reddit is a very smart audience. They hate marketers. They hate being sold to." (Founder, 07:07)
Posts range from zero upvotes to three or four hundred, so he judges them on traffic and conversions (Founder, 10:39).
What is the start-over playbook?
Reddit:
- List at least 15 communities where your buyers spend time (Founder, 08:19); anything over 5,000 members is fine, and small ones are often better (Founder, 08:34).
- Pick one feature or pain point, make a short video demo, explain why you built it, and offer a 30-day free trial with tracked links (Founder, 09:02).
- Tweak the copy and repeat in one new community per day, never all at once (Founder, 09:29).
Search:
- Know your exact position in the market first (Founder, 13:47).
- Filter keywords to difficulty under 20 and volume over 500 (Founder, 14:14).
- List 30 to 50 topics, then publish two or three posts a week, scaling to six or seven (Founder, 14:41).
Why does it win?
He says users trust the answers because they come only from the documents they added, not from a general chatbot that may make things up (Founder, 16:37). It also plugs into tools people already treat as a second brain, such as their notes app (Founder, 17:04). The interview does not compare it with free or better-funded rivals.
How is it priced, and why?
Three standard plans: monthly, annual and a lifetime deal. Mac app buyers expect a one-time payment, so they have kept the lifetime deal from the start and raised prices gradually (Founder, 01:46). The interview gives no plan prices and does not say what the price rises did.
What does it cost to run?
An AI coding assistant at $100 a month, used for both code and marketing, and a keyword research tool at $129 a month (Founder, 17:31); hosting, email and analytics tools are named without prices. The interview does not cover margins or salaries.
What went wrong?
He and his partner had more than 30 failed projects before this one (Founder, 19:27). On Reddit they got hate and bans and learned to live with it (Founder, 10:11), and moderators eventually removed their posts, cutting engagement (Founder, 07:33).
What is the founder's advice?
Put more buy buttons on the internet instead of building in silence for months behind a waitlist (Founder, 18:54).
"I think people are too afraid to charge, too afraid to put up a real buy button." (Founder, 18:54)
Then test many ideas in short sprints and double down on the few that earn (Founder, 19:27).
What did the hosts take away?
The host's point is that Reddit's value is the honest feedback, not the traffic:
"It's about the engagement and the comments and the feedback you get from real people." (Host, 20:49)
Many founders show him a first post with about 50 upvotes, and anyone expecting $10,000 from one post has it wrong (Host, 21:17). The second host's takeaway: the playbook still works and costs only effort (Host, 20:22).
What to copy
- Use each channel for what it is good at. Communities give feedback and first customers fast; search compounds later. Switch when the first one flattens.
- Treat support docs as search content. Answers to real customer questions can rank before you ever do keyword research, and they show you what to write next.
- Ask for money early. A real price on a real page teaches more than months of building behind a waitlist.
More from this series: free tools as an SEO engine, the first 100 customers from Reddit, every Starter Story episode answered, and our free YouTube transcript tool if you want to run the same questions on any interview.