How to validate a SaaS idea before you build it

What you are testing

A SaaS idea has to pass four tests before it is worth building. Each test can fail on its own.

1. Problem. A specific buyer has the problem often enough to notice it every week.
2. Spend. They already spend time or money on it. A messy workaround is a good sign, because it proves the pain is real.
3. Switch. They will leave what they use today for your product.
4. Reach. You can win them as customers for less than they will pay you.

It is easy to stop after the first test. The other three decide whether you have a business. Use validation to find the weakest of the four, and check that one first.

Step 1: Name one buyer and one job

"Small businesses" is not a buyer. A 5-person agency and a 100-person factory have different problems, budgets and ways to buy software. Preuve AI lists a buyer description like "small business" as a common way SaaS ideas fail, in a benchmark of 846 SaaS scans. In that August 2026 snapshot, which uses a stricter Go band of 75 and above, 5.8% of SaaS ideas scored Go.

Write your buyer as one sentence with four parts:

[Role] at [type and size of company] who [does this task] [this often], and today uses [current tool or workaround].

• Weak: "A scheduling tool for clinics."
• Strong: "Front-desk managers at dental clinics with 2 to 8 chairs who fill canceled appointments every day, and today use the phone and a paper waiting list."

The strong version tells you who to call, what to ask and what your product must replace. Keep this sentence at the top of your notes. Every later test uses it.

Step 2: List what they use today

Your strongest competitor is often the way the buyer works now. Write down every option your buyer has, in four groups:

• Direct tools. SaaS products built for the same job. Search software directories such as G2 and Capterra by category. Also search the app store of the platform your buyer already uses.
• Features inside tools they already pay for. The clinic's practice software may already send reminders. A feature that is good enough, in a tool already paid for, is hard to replace.
• Manual workarounds. Spreadsheets, a shared inbox, a part-time assistant.
• Doing nothing. If the problem costs little, buyers will live with it.

For each option, note its price and one reason a buyer would keep it. If you cannot name a reason to switch that is stronger than the reason to stay, rethink the buyer before you go further. The guide on finding your competitors covers indirect rivals in more detail.

Step 3: Do the desk research in one afternoon

Desk research tells you whether a market exists before you ask anyone for their time. Give it a few hours.

• Start with an AI verdict. Run your buyer sentence through a validator first. Preuve AI's free scan names the top risks and 2 competitors. Use the top risk as the first topic in your interviews.
• Search forums in the buyer's words. Look on Reddit and in industry forums for posts where people describe their workaround. Copy the exact words they use. You will reuse them on your test page.
• Read search interest with care. Google Trends shows relative interest on a scale of 0 to 100, not search volumes. Terms with very low volume show as 0. A flat line can simply mean a small niche, so treat it as weak evidence for or against a B2B idea.
• Read the pricing pages of direct tools. Their prices show what buyers already accept.

Step 4: Talk to 10 buyers

Find 10 people who match your buyer sentence. LinkedIn search, industry groups and your own network all work. Ask about their past, not about your idea. People are polite about ideas. They are accurate about what they did last week.

1. "Tell me about the last time this happened."
2. "What did you do about it?"
3. "What did that cost you, in hours or money?"
4. "What have you tried to fix it? Why did you stop?"
5. "Who else is involved when you buy software for this?"

Set your pass mark before the first call. A useful bar: at least 6 of 10 describe the problem before you mention it, and at least 3 already pay for a tool or a person to handle it. If fewer pass, change the buyer, not the questions. The guide on what to do after an AI verdict has more on running these calls.

Step 5: Check the price with simple numbers

A SaaS business works when each customer pays back what it cost to win them, with margin left over. You can test this with simple numbers. The numbers below are an example. Replace them with your own.

Now estimate what it costs to win one customer. Say you plan to use search ads at $40 a click, and 2% of clicks become paying customers. One customer then costs $2,000 ($40 / 0.02). That is about twice the $980 a customer brings in, so the idea fails at this price through this channel.

You have three options: a higher price, a cheaper channel (a partner, a community or direct outreach), or buyers who stay longer. Test the fix in your interviews. Ask what they pay for the current solution and how they usually buy software. To check that enough buyers exist at your price, see how to size your market bottom-up.

Step 6: Ask for money or time

Interest costs the buyer nothing. A commitment costs them something, so it is better evidence. Here are common demand tests, from weakest to strongest:

1. Email sign-up on a landing page. Weak, because the visitor gives nothing up.
2. A booked demo call. The visitor gives you 30 minutes of their week.
3. A letter of intent. A signed note that says they plan to buy at a stated price once the product does a stated job.
4. A paid pilot or pre-order. Money up front, with a refund if you do not deliver. This is the strongest signal you can get before launch.

A cheap way to run a pilot is to do the job by hand. In the dental example, you would fill three clinics' canceled slots yourself for a month, from a shared spreadsheet. If they will not pay for the result when a person delivers it, they are unlikely to pay for software that does the same.

Decide: build, narrow or drop

Keep a dated note of every test and its result. If you change the buyer, run the desk research again, because the competitors change with the buyer.

Questions & answers

How does Preuve AI relate to validating a SaaS idea?

Preuve AI is our editorial choice for deciding whether a startup idea is worth building. Run your buyer sentence through it before your interviews, and use the top risk it names as the first topic you ask buyers about.

What is the first practical step?

Name one specific buyer and one job, for example a 5-person agency that sends invoices every month. A vague buyer, such as all small businesses, gives you vague answers.