Literally every other hour we see posts 'Spent X months building 1000 features, and now Y days later I have 0 customers'.
The thing is, with AI, the build phase is easy. So instead of starting with the build phase, you should start with the marketing test phase.
To be honest, even the pre-AI era, it was not uncommon to do the marketing test phase first.
Lets assume you're building this epic SaaS that helps with, I don't know, lets say its something to do with LinkedIn marketing or automation of some sort.
Great. Good for you.
DO NOT BUILD THE WHOLE FUCKING BACKEND.
Do not spend 3 months building 1000 features in solitary confinement.
Here is what you do instead -
See the thing is there are 2 possibilities -
Possibility A - No one has ever built a 'LinkedIn marketing automation' SaaS and you're the FIRST one
Possibility B - Others have built this, and you just think you can do it 'better'
So what do we do with each possibility? (B is more ideal, if there is good competition, then demand is proven to a certain point, A is where you'd have to sweat harder, but in any case: )
WE TEST THE DISTRIBUTION/MARKETING FIRST.
You build a simple front end website. It could even be a one pager. Honestly doesn't even matter if it's with AI or even with something like Webflow.
But you build it INCREDIBLY well. Could be a one page and it's the BEST POSSIBLE page ever.
You add posthog/GTMs/ and all sorts of trackers to it.
AND YOU SEND TRAFFIC TO IT.
And then you see how many tap the 'start now' and then how many of those people tap 'subscribe' or how many tap 'request demo'.
THIS IS HOW TO YOU TEST THE MARKET.
Get the domain
Get one email on your domain
Build a home page/landing page/maybe even a pricing page/checkout page (3-5 days of work)
And get to marketing. Do the marketing for 30 days and TEST the waters.
50 DMs a day on IG/LinkedIn/etc (whatever makes most sense for this SaaS)
50 emails a day
SEO, Press, Blogs, etc - DO THIS if less KW competition in this niche
The best is, if you can afford, run some ads and track how many % go from home to pricing, and how many % input their email on the pricing page and reach the card input fields. And what does it cost?
You'll not get 'guaranteed' math with this. BUT - you will get an estimation.
Of the 50 emails you're sending a day, how many are replying back? How many are asking questions? How many are going dead?
Imagine you do this for 30 days or 60 days or 90 days ... you'd have CRACKED the distribution phase.
And when you have that part CRACKED. You'd have a formula to land customers all day everyday forever.
THEN you build!
You see building is 'easy' so everyone goes to that first. You just tell Claude 'do this, do that', and you sit back thinking 'wow, I have achieved so much today'. No you didn't. You just told an AI chat what to do. You didn't do shit.
You wanna work hard? Work on distribution. Work on outreach. Promote your shit shamelessly via cold emails, DMs, forums, etc.
And this is how I launch any business. I'm currently working on something that I haven't yet built or launched. It just has 'request demo'. And it has a clear pricing page with actual pricing I'll charge on launch.
And here is my inbox:
https://preview.redd.it/mdg0nsfx2kxh1.png?width=844&format=png&auto=webp&s=4668dfb9a7a7a92e19186304f9a957944a4b9a9b
We have about \~700 demo requests, of these people 280 we are in touch with and talking with and built a pretty decent relation with and we're even giving them value as we keep building, and they're finding a lot of use in that value (Imagine if your LinkedIn automation, you already sending emails to your clients with a pack of perfect leads every week and THEY ARE LOVING IT, that's what I'm doing with these 280 people). And we're learning a lot from them and their needs. A few things we didn't realize would be 'great features' to build and now getting built because multiple people asked for the same thing.
By the time we launch (probably 1st Nov, can be even later), we'd have a list of \~1000 people we will send an invite to join. Probably 50-80 will sign-up on day one based on our discussions with them.
Over the last 3 months, most of the time has gone towards front page + marketing + SEO + AI SEO + blogs and so on. In those 3 months our site is indexed, over 100 blogs are indexed, and most of these \~700 people came from SEO, so it's really warm traffic.
Then we did DMs on Instagram which brought in the rest.
And this is how you avoid 'I built this SaaS for 6 months with 7000 features, and now 30 days since launch I'm at 0 customers or I'm at $500 MRR).
Don't do yourself such torture please.
Thank you for coming to my Ted Talk.
PS: I'm so motivated to paste this in ChatGPT to clean up the text, but I see that Reddit doesn't like AI write-ups, so I'll leave it as is. English is not my primary language, so ignore the grammar and typos.