[Fatboyrunning]

Sorry but I am not going to pay for this before trying it out on my content.

Surely there can be a test function within the website which allows me to see what a user would experience?

[mpaepper]

You can do it yourself with this repo and only pay the OpenAI API fees: https://github.com/mpaepper/content-chatbot

[Veen]

Your content-chatbot repo was very useful when I was figuring out how to achieve this sort of thing with langchain. I was able to knock together a chatbot for a client's documentation site in an afternoon. But I guess the real value for SiteGPT is the ease-of-use and the client-side chat interface.

[pqdbr]

Same here. I know it hurts to offer a free trial for something that already costs you money to serve (those API calls won't be free), but it's really hard to sign up without trying it first.

[pbteja1998]

Yes, I understand. I will give you the same option as the previous person. Please give me a sample webpage, just one web page and I will create a chatbot for that webpage, and post the chatbot link here.

[marktani]

I wonder if there is a way to let users generate this one-page demo themselves dynamically, rather than via a manual interaction by you.

what's to say you won't add specific optimizations by hand now that aren't included in the product out of the box?

edit: ah, just saw your other comment about being afraid free trial gets out of hand financially

a solution could be to only allow x number of free tokens for a given demo chatbot

[pbteja1998]

I understand. In that case, the only way is to subscribe to the $19/month plan and try it out for yourself. You can cancel anytime you like with the click of a button.

[pbteja1998]

Yeah, I didn't know how much that is going to cost me if I added a free trial. But I can create a demo for you if you like. Give me an example web page link for which you want the chatbot.

I will create one and post the link here. Just a single page url.

[mpeg]

I don’t mean to be harsh, but if running a free trial would bankrupt you, you shouldn’t be trying to start a company.

It also doesn’t inspire much confidence in your early users, there’s been a lot of these GPT API cashgrabs popping up all over so if you want to differentiate yourself you might need to actually incur some risk.

[dorilama]

If you can't risk 19$ to see if a service can bring value to your website maybe you shouldn't give advice about starting a company?

[mpeg]

I think I'm qualified to give this advice, seeing as some of the biggest brands in the world have trusted my advice on digital marketing.

It has nothing to do about whether I personally can risk the $19, I'm not even in the target audience for this – the question is what percentage of the target audience is going to be ready to pay $19 for something they don't know is going to work for their site, and how much bigger would that pie be if the site owner spent a tiny amount on offering trials.

Just making people get their card out is going to make a huge percentage of leads drop off, especially when there's almost no content/demos or an actual working trial in the site (even the screenshot is just a static screenshot instead of a live demo).

If you want to know more, you can reach out via email and I'd be happy to help (though it might cost you a bit more than $19)

[phil-martin]

What if instead of offering free trials to everybody, you estimate how many free trials you could afford, and limit it to that?

In exchange for the free trial the potential customer gives you permission to use it as a demo for others to try out and see how it performs.

[adrr]

Same boat. Not paying unless you can show me a demo of it working with my content. Do even know if their scrapper can properly scrape our site.

[robertd7]

You can try out https://chastfast.io. Doing same thing

[robertd7]

Edit: https://chatfast.io

[r0fl]

Exactly what I’ve been looking for

Great job!