Looking for advice – product, idea, marketing or website?

Hey everyone,

I launched my product about a month ago (after a beta period). It’s an AI business mentor that helps early-stage entrepreneurs (0–3 years in business) get clarity, guidance, and actionable support for growing their business. But I’m struggling to get active users. I’d love to hear your feedback on whether you think the issue is more with the idea, product itself, marketing strategy, or the website/funnel.

Here’s what I’ve tried so far:

  • YouTube Shorts: ~16k views in the last 28 days (~180 views per video, posting 3 per day)

  • TikTok: ~20k views in the last 28 days (~230 views per video, same 3 video per day)

  • Reddit: posts usually get 300–350 views, comments on relevant threads ~10 views each

  • Twitter & Instagram: tested but didn’t see much traction, so I didn’t continue

  • Product: live for 1 month, with some free signups, but no one is using it actively yet

:right_arrow: Website link: Voltrex AI

I’m wondering: do you think the bottleneck is more likely the idea/market fit, the product itself, my marketing channels, or the website? Or maybe these are good results and should I just continue what I’m doing now, always improve a little and it will start working, I know that 1 month live is a very short time, but I don’t want to start in the wrong direction

Any advice or honest feedback would be super appreciated. Thanks in advance!

(If you’d like to test it in more detail, let me know and I’ll provide full access.)

P. S. Yes, of course I asked my AI, but I’m curious to see what a real person would say and compare it to the answer I got from the AI.

1 Like

Hey James,

I’d suggest that at this stage you want to do some usability testing where you sit down with a target user, get them to use the app in front of you and give you feedback as they use it: what they like, what’s confusing, what’s missing. This allows you to get some real user data to help prioritise what to do next.

Marketing it through social media is going to give you some quantitative data (e.g. this many people hit the landing page, this many people activated the app), but it’s not going to tell you why they did or didn’t use it. So it will tell you that there is a problem, but no insight in how to fix it.

I had a go at using your app here are some initial thoughts/ideas:

I’d suggest making the hero section on the landing page clearer. Build and grow your business is quite high level, reading it doesn’t make it any clearer what the app will actually do for me. I’d also think about including some visuals of what kind of output I’m going to get. I’d like to see something that made me think “oh that would actually be really useful for me”. At the moment the guy in the hamock doesn’t really align with your value proposition. Also, my initial thought is “why can’t I just do this in chatgpt?”, so it might be worth making it clear how your app differentiates itself from chatgpt? e.g. trained on some data that openai doesn’t have etc.

So I’d say on the landing page their isn’t enough clarity on what value the user will get to help them make the plunge to using it.

On the onboarding there is a lot of friction before you get to anything valuable. I wonder if you could jump soon to the AI part - there are a lot of inputs and drop downs, which feels quite traditional, I wonder if you could run the question in the chat interface? you can create sequential agentflows in flowise or n8n where you can set it up so the user can’t graduate to the next question without answering the previous.

Once you’ve completed the onboarding questions you need to create and account, then verify the email address. If the landing page had made me super excited about what’s on the other side, I’d probably be ok with doing this. But otherwise, by this stage it feels like a lot of work, when you’re not sure what you’re getting.

Hope that helps.

Ryan

1 Like

You have hit the hard part of being in business. There is a reason that most people work for other businesses like agencies, freelance marketeers, content creators etc. Because a business with a new type op product, without many followers, large network or authority is very hard.

This has to do with the fact that humans don’t like to start something that is out of their playbook. They rather do what their friends do or some popular figure. Just to make sure they don’t get laughed at or prevent them from a mistake.

With that in mind, you have to look at your product. How does it handle the above objections? Does the landing page give them trust? How about the url? How about the social proof? How much time do they need to invest to get some notion of the value? Same as for the distribution channels. Who is vetting for your product?

Another important mindset is the difference between product thinking and market thinking. The biggest challenge is almost never with the product or the technical part for that matter. That is relatively easy. The hardest part is to get it in front of real users and sell it. That’s what you experience now as well. You have to embrace it, feel it and act accordingly.

Some other tips:

  • Sell emotion and not features. “Outsmart your competitors and make sure you do not tell them about your secret unfair advantage” you get the idea…
  • People always want to be seen, why not use AI to start a conversation and forget about the form filling. Get the conversation going with a simple cheap ai model and store the answers and gradually handover to more expensive models later in the process. Forms and questions feel nothing like AI or premium to me for that matter.
  • Target a very small group of possible first time users. Who are they? What are their issues? What do they need help with. You can focus the product much better so when such a user hits the product it immediately thinks “this AI thing seems to know what I am dealing with!”
  • Think about your own journey. What do you struggle with now and why do you go to a forum like this and not use your AI product yourself?
  • Friction, you can have the best product in the world but people are lazy, they don’t read things, they don’t want to try things so you must make sure there is almost no friction to get to value with your product. That does not mean that it needs to provide clear answers to complex business problems out of the blue. It means you have to give any type of value quickly with very little effort from the user. In same cases this means that the user has to do more then necessary btw. I someone is used to WhatsApp, don’t ask them to use telegram. If they are in TikTok, don’t let them go to a website. The point, the interface they use might be less than ideal for your product but they know the interface so less friction to use it. Show them, teach them and eventually they will figure it out themselves.
  • Trigger, almost everyone forgets to think about the trigger. When does you ideal customer thinks “I wish there was this product that allows me to…”. You have to present yourself at those moments and at the places where the customer is experiencing those moments
  • Alternative, think hard about alternative ways your perfect customers solved the issues you solve. Phone call — letter, car — train, feeling cool — wear product x etc etc

I hope this helps. I know first hand how hard the journey is from an idea to paying customers. If you believe in it and are passionate about it, just do it! If it is about making money and money is not the most important thing in your life, stop it.

1 Like

Thank you guys, all the tips, I will definitely use them, really appreciate it

And do you guys think it’s a good idea? Or you wouldn’t pay for it and use chatgpt instead, even if it’s not that great. (What would be that one thing, you would pay $30/month)

I would use ChatGPT but that’s because we do so much with it, it feels natural to do so.

I would consider something that delivers vertical value. Most products, even with AI, deliver horizontal value. But that will be easy to handle for llm models in a few years out of the box I think. Vertical is much harder, and for AI even exponentially harder.

1 Like

Thanks, good point