Revenue Isn't Everything in Early-Stage SaaS – What I Learned in the Past 8 Mo

6 FlorinLazar 4 6/26/2025, 10:25:04 AM
I've spent the past 11 years working in startups, from launching my own outsourcing software company to now building Assista AI, an early-stage SaaS helping non-tech professionals automate tasks using natural language.

I've heard many things about raising funds, but one remains consistent. Fundraising conversations often come down to a single question: "How much money are you making?"

But recently, I've realized focusing solely on revenue at the earliest stage can be unhealthy, even harmful.

Here's why:

When revenue is your primary early focus, you're pressured into quickly building something sellable, but not necessarily valuable. Truly great products usually need iteration, feedback loops, and experimentation. Immediate pressure for revenue often robs you of the space to refine your product.

I followed this pattern many times until we approached things differently at Assista AI (https://assista.us). I'm an engineer by degree, so I believe in process.

Therefore, we split the conversion process into 4 steps that worked well for us recently:

1. Marketing: Identify your audience and their needs first.

From the first users, we quickly learned our users were non-tech professionals from English-speaking countries. Then, we experimented with many channels with small budgets to see what would be the best place to acquire our users:

- We assumed Influencers via affiliate programs could be powerful for us. It proved right: we brought 500+ users at $200 total. - Reddit was great for awareness (800 landing page views at $250), but not conversions. - Google Ads brought in targeted users, but at a high cost ($21/user). Not for us at the moment. - LinkedIn organic has been promising due to our professional target audience. Downside: It needs a lot of time to create valuable content.

Overall, good results and a lot of learning. We know where to focus next.

2. Landing Page Conversion: Once they land on the website, do they create an account?

One month ago, we operated on a freemium model. We saw many accounts created (around 15% landing page conversion rate), but with low interaction. So, it meant we had a good message, but many people weren't part of our target audience. So we moved to trial. As expected, the conversion rates dropped from 15% to 5% but...(check next)

3. Usage:

..session length grew from 2-3 minutes to 10-15 minutes. People introducing their card to access a trial is a strong signal that they are willing to pay.

Also, longer sessions mean more data to analyze, and compared with freemium users, the ones on trial will share their feedback, and it's not always positive. At the end of the day, they add their card, so they have expectations. That's normal.

4. Trial to Paid Conversion: This is our current focus, understanding what delivers enough value to convert trial users into paying customers. Even if the product isn't perfect (we still have many bugs) and is far away from what we envision it can be, we see strong signs we are on the right path. We will have more results on this in the upcoming weeks.

Meanwhile, we got our first pre-seed investment, so we have more money to experiment, and we are raising our seed round, which will move us to the next chapter in terms of marketing and product.

Conclusion:

Treat your SaaS as a pipeline where a paying user is the last step. After we find the answer for the final step, there will be one more: churn.

Work from top to bottom and improve one step at a time. If you improve each step, revenue will come naturally.

Today you test smth on marketing and you see how it goes. Tomorrow, you will analyze their behavior on the app. Then iterate on the marketing and the app. For us, 2-week sprints work best.

What I wrote about is our experience. It's not perfect, but if you take one valuable idea from it to implement for your project, I would be happy.

Comments (4)

yashasolutions · 7h ago
That's what the concept of lean startup is about.
FlorinLazar · 7h ago
Thanks! This is the whole idea. You summarized it very well. However, bootstrapping like this until receiving first funds requires a lot of effort, determination, and a team that understands the process. We are lucky to have it.
m3d1n · 8h ago
Great read! How do you decide which feedback to act on?
FlorinLazar · 8h ago
Thanks! Well, that's a good question. What we are looking for is to correlate quantitative data from our databases with qualitative data by understanding users' behaviors, and with feedback.

So, feedback itself is valuable, but it needs to be linked with the other input sources to be prioritised. Otherwise, we thank users for their feedback, but if it's not smth that we receive enough times, we don't act on it.