My role: technical partners write codes. I'm in charge of growth
Elephas was not the first one i started alone。
My partner kamban did the product first, when i just resigned from 11 years of company work and was working as a product and marketing consultant for some of the early saas companies。
I started as a consultant. The first five to six months were good, and he later invited me to be a co-founder, full-time in charge of markets and growth. He's in charge of the development. I'm in charge of bringing the product to the user。
That is why i am paying particular attention to the growth approach. Many people now use the ai tool to make the product, but when it does, the real problem begins: where does the user come from? Where did the first feedback come from? How can more high-intensity users find you
For us, first reddit, then seo。
Phase 1: 3000 mrr from 0 reddit
Reddit is a high-intensity gold mine broken down into users。
It suits the early product because you can quickly find the target user, get real feedback and get the first flow and transformation。
In the early days, we posted in many of reddit's subdivisions on the theme communities, asking users to provide feedback, display products, and do functional presentations. These posts bring in the first users and let us know where the product should be changed。
Shit, reddit, we've got $3000 from 0 in six months(the president)。
But reddit has borders. It is well suited for early validation, but the more it is invested, there will be marginal gain erosion. You can't grow every day by posting。
So we started turning to seo。
Reddit? How do you post it? Not like marketing
There's a kind of post that works well for us: a video presentation around a specific issue。
It's probably written:
I or my friends have this problem, so i did this function. Here's how it works. If you want to try, you can try for 30 days for free. I'd like to hear feedback mainly: where can this function be changed
This expression is valid because it is not hard。
Reddit users are smart and hate being marketed. They don't like you coming up and saying, "how good is my product?" but what do you think if you explain it like a normal person: why do i do this, what does it solve? They are willing to discuss it seriously。
Video presentations are important. It's show, show (not empty description). Let's not just say it's a good product. Let the user see what it can do。
We just introduced superbrain, and we don't know if it's working. Following the posting, we received a large number of ongoing feedback and functional requests. It became the primary function of the product, and today many users have used it to index their own information, chat and generate content。
If today starts at zero, i'll do it, reddit
First step, list the subreddit where the target user is located。
Find at least 15. Open-source tools such as map of reddit could be used to expand from a known community。
There is no need to focus only on the large community. The community of over 5,000 people is worth trying. Small communities are sometimes better off because users are more vertical, rules are simpler, and owners prefer valuable content。
The second step is to select specific pain points or functions that a product is addressing and to record a short video presentation。
Explain in very simple terms why you did it and send it to a subreddit. At the bottom, a product link can be used to write “30 days of trial free of charge, welcome feedback”。
The link is to add utm parameters (link tags to track the source of traffic), so you can know which day the post will be, which subreddit, which function it will be around, and how many visits and transformations it will bring。
The third step is to send only one community a day。
Don't send the same text all day long. Many community users overlap, and you can see at once that you're painting. One subreddit a day, then fine-tune the text based on the feedback and send the next。
And for 14 to 15 days in a row, you get a lot of data: which headings work, which functions trigger discussion, which communities are better suited to your products。
Reddit's success criterion is not a compliment, but a real feedback
We have zero posts, 30, 40, 50 general posts and 300, 400 posts。
But i don't think praise is the only criterion。
What really depends on flows, transformations, and qualitative data in the reviews. How does the user understand your product? What are they confused about? Do they think it works or not? What are they willing to argue about
Reddit's worth right here. It is a place where a few people can still get a lot of real, direct and even hard-on feedback。
And, of course, you'll have negative comments, and you'll be cut or blocked. It's part of the game. Business takes shame. Negative feedback is also training in itself, which will enable you to learn to face real markets earlier。
Phase 2: seo pushed us from $3000 to now
Reddit helped us get early users, but seo made growth more stable。
We actually found seo by chance。
Early users asked a lot of questions and we wrote some support articles to help them. For example, "how to create openai api key." the article is simple, basically a screenshot, links and steps。
And then we found that it was number one on google, and it lasted six to eight months and brought in a lot of traffic。
Although this type of article may not have a strong intent to buy it, it reminds us that google is a real access channel。
In 2023, we made about $70,000 from google natural flow. And it's seo that brought us from $3000 to today's level of about $120,000 a month。
Then we started to do more systematic keyword research, looking for topics that are low-competitive, high-demand and more relevant to products。
One of these works is “18 best chatgpt mac apps free and paid”. This type of checklist directly corresponds to the purchase intent of the user: he's already looking for the chatgpt mac application, just don't know which one to choose。
This article is well ranked and brings with it a lot of high-intensity traffic and transformation. Now, not only does google bring traffic, but chatgpt, claude, an ai ai answer engine like this, will quote our article. Users coming through the ai answer engine tend to be more intent, as they are close to purchasing decisions。
If we do the seo today, i'll do the location first
Seo's first step is not to write, but to clarify where you are in the market。
What do users think about the market? What kind of word do they search for? Which subcategory do you belong to
For us, the big class is chatgpt apps, and more specifically chatgpt mac apps. In this small market, users will have different applications and find tools for themselves。
Once clearly located, the keyword study makes sense. You're going to think from the point of view of the target client: what would he search for in google? What are you gonna ask in chatgpt? What are the key words of the competition
I'll do a keyword study with ahrefs. A practical filter condition: keywords are less than 20 and search volumes are more than 500. For the new site, no high-flow, high-competitive words should be taken from the outset. A better strategy is to find low-competitive, high-intensity words。
Before ahrefs was $129 a month, it wasn't cheap for us. I'll buy a month, concentrate on 500 keywords, sort out 30 to 50 themes, and write slowly in the next few months。
It would be sufficient to start writing two or three times a week, which would be raised to six or seven times a week。
Ai is available but does not allow it to mechanically rewrite existing content on the internet. The better way is to use ai as a research assistant to help you analyze which articles are effective and then add your own insights, user data and market experience。
The real focus of seo is: can you add new information to the internet。
It's hard to win if it's just a repetition. There is an opportunity if you bring your own case, user feedback, product data and field experience。
Our toolhouse
Elephas is a native mac application developed with swift。
Both development and marketing uses claude code extensively, approximately $100 per month。
Seo research using ahrefs, $129 per month。
Before the blog is published, check whether the article is complete compared to the competition in the search results。
Reddit research map of reddit。
Notifier, we can respond in time once someone mentions elephas in reddit or other social platforms。
Team task management for clickup, communication and user community for discord。
Web hosting is based on superblog, maillite for mail marketing, plausible for website analysis, and google seach console for seo data。
There are many tools, but the bottom logic is not complicated: reddit is responsible for early feedback and for the first users, and seo is responsible for bringing in people who clearly need them。
Last suggestion: add a few real purchase buttons
If i had a suggestion for early entrepreneurs, i would have said, put more purchase buttons on the internet。
A lot of people do six months of silent production, put on a free waiting list and expect the users to come naturally. No one bought it, and he didn't know what the problem was。
The real purchase button will bring a completely different learning。
When you release the product and let the user pay for it, you will know whether they hesitate, ignore or buy the card. This feedback is more direct than any marketing book, sales class。
Of course, you have to fail many times to get there. Me and the partners did more than 30 failed projects before elephas。
So my suggestion is to consume the failure as soon as possible。
Multitesting ideas, short-cycle sprinting, quick failure, quick learning. Then we look at what really pulls, what really gets, and then double our energy。
Reddit and seo are not magic. They just make you face real users faster and let those who really need you find you。
