A SaaS website not converting is, in nine cases out of ten, a page that answers the wrong questions. Here is a quick way to see it. Open your homepage and read it as a stranger who has thirty seconds to spare. Then ask what question each section answers. On most B2B SaaS homepages, the sections answer questions the founder had, and they skip the questions the buyer has. The founder's questions sound like this. "What did we build?" "What do we believe?" "What can it do?" A team asks itself those questions during a launch. A buyer does not bring them to the page.
The buyer brings five other questions instead. They ask them in a fixed order, and they ask them in plain words. They leave at the first question the page does not answer. This post lists the five questions, shows what the page usually says instead, and gives the fix for each one.
Why the page answers the wrong questions
Nobody sets out to write a homepage for themselves. It happens because the people writing the page know the product too well. The founder knows who the buyer is, so naming the buyer feels like a wasted line. The team knows the product works, so adding proof feels like bragging. Everyone knows what the product does, so the page skips ahead to how it does it. The result is a page that makes sense to the people who built it and says nothing to the person it is for.
The second cause is fear of leaving someone out. Naming a buyer turns some visitors away, so the page says "teams". Naming a price loses some deals, so the page says "contact us". Naming a specific result can be proved wrong, so the page says "Grow faster". Each of those choices removes an answer, and answers are what the buyer came for. We wrote about the root cause in positioning before pixels. This post makes the same argument from the buyer's point of view.
1. "What is this?"
The first question is the simplest, and it is the one most often skipped. The buyer wants a category. Is this a CRM (a tool for tracking customers), an analytics tool, a signage platform, or an agency? Until they know what kind of thing it is, they cannot judge anything else on the page. They have nothing to compare it to.
What the page usually says: a benefit with no noun. Examples: "Better decisions, faster." "Work smarter." "The future of customer engagement." Each line describes the result of using something, without saying what the something is. In our SuperOps teardown, the largest text on the page was a statement about the age of AI. The category appeared only in the meta title. The meta title is the page name that shows in Google results. A visitor who is already on the page never sees it.
The fix: put the category noun in the H1 or in the line directly under it. The H1 is the main headline at the top of the page. Examples: "Digital signage software." "IT operations platform." "Sales assistant." These are plain words, they match what people search for, and the buyer can place them at once. If your category is new, name the nearest familiar category first. Then say how you differ from it.
2. "Is it for me?"
Once the buyer knows what the product is, they check whether it is for someone like them. They look for company size, role, sector, and the tools they already use. They are scanning for one word that matches their own situation.
What the page usually says: lines written to include everyone. "For modern teams." "For businesses of all sizes." "For anyone who wants to grow." Each of these matches no one. A buyer looking for "for MSPs" (managed service providers, the IT firms that run other companies' systems) does not find that phrase. The same goes for a buyer looking for "for Series A SaaS" or "for retail chains". They assume the answer is no and leave.
The fix: name the buyer in the first screen. If you have one buyer, put them in the main sentence. If you have two or three, give each one a card with their name and a one-line promise. Put the cards directly under the hero, which is the first block a visitor sees on the page. Pickcel moved from feature-first marketing to separate stories for retail, QSR (quick service restaurant) and enterprise buyers. After that, procurement teams stopped asking the same questions on every deal. The page had already answered those questions for that buyer.
3. "Does it work?"
The buyer now knows what the product is and that it is for them. The next question is whether to believe the page. Enthusiasm does not help here. They want something they could check on their own. That could be a customer they recognize, a number with context, a case study they can click, or a review on a site they trust.
What the page usually says: "Trusted by industry leaders." A logo wall with no context. A testimonial from "Sarah, Marketing Manager" with no company name. An "as seen in" strip. A skeptical buyer cannot verify any of these, and skeptical buyers are the ones with budget.
The fix: put one checkable proof point in the hero, next to the main sentence. For example, "Runs the screens in 4,000 stores across 40 countries", if that is true. Or use a named customer with a linked case study and a number in the title. If the numbers are not ready yet, a named customer alone beats an anonymous quote. If you have no proof at all yet, say what the product has done so far and for whom. A small proof point that a buyer can check beats a large one that nobody can verify.
4. "What will it cost me?"
Cost means more than price. The buyer is asking what the product will take from them. That includes money, time, integration work, a migration, and a conversation with procurement. They want a rough answer before they commit to a call, because the call itself costs them time.
What the page usually says: nothing, or "Contact sales". Sometimes there is a pricing page with three columns and no guidance, so the buyer cannot tell whether the middle tier is for them. The page rarely says anything about time: how long setup takes, what has to change, and who has to be involved.
The fix: state the shape of the price even if you cannot state the price itself. "Per screen, per month, from" is a shape. "Annual contracts, typically five figures, priced per seat" is a shape. Then answer the time question in one line. "Live in a week" or "A two-week onboarding with our team" both work. Buyers do not need the exact number to keep reading. They need to know that your pricing is in the same range as their budget.
5. "What do I do next?"
The last question is the one the page was built to answer. It is also the one most often answered in five different ways at once. The options look like this: "Start free", "Book a demo", "Talk to sales", "Watch the video", "Read the docs" and "Download the guide". Each one is a reasonable next step on its own. Together they force the buyer to make a choice they did not come to make.
What the page usually says: all of them. In the Mutiny teardown we counted eight competing calls to action on one page. A call to action is a button or link that asks the visitor to do something. Each of the eight made sense to the team that added it.
The fix: offer one primary action for the buyer you named in question two. Offer one secondary action for the person who is not ready yet. Move everything else to the navigation menu or the footer. If you sell to two buyers, the two cards from question two can each carry their own next step. Here is the test. A buyer who has said yes to questions one through four should not have to think about what to click.
The questions the page answers instead
It helps to name what fills the space where the five answers should be. Each of these feels like useful content while the team is writing it.
- "What can it do?" The feature list. It is useful on a product page. On the homepage it arrives before the buyer knows whether the product is for them.
- "What do we believe?" The mission section. Buyers read this after they trust you, and they skip it before that.
- "What is it built with?" Words like "AI-powered", "cloud-native" and "enterprise-grade". These describe the engineering. They sit where a claim about the buyer's result should be.
- "Who has noticed us?" Awards, badges and press logos. These prove that someone else looked at the product. They do not prove that the product works.
- "What is new?" The launch banner. In the SuperOps teardown, a private beta announcement was the first thing on the page. It sat above the product the buyer could actually buy.
None of these belong in the first screen of the homepage. They are fine elsewhere on the site. In the first screen, they answer questions that nobody arriving at the page has asked yet.
A ten-minute test
Open your homepage. Write the five questions on a sheet of paper. Next to each question, copy the single line from the homepage that answers it. Quote it exactly as it appears. Do not write the line you wish were there.
If any of the five questions has no line, that section is costing you inquiries. Fix that one first. If a line exists but sits below the fold, move it up. Below the fold means the part of the page you only see after scrolling. If the answer to question five runs to more than two lines, cut it down. Then leave the design alone for a month. Watch the number of inquiries, and watch their quality even more closely. When the page answers all five questions, the wrong buyers stop writing to you. The right buyers arrive with their problem already described.
The full section-by-section version of this advice is in the B2B SaaS homepage guide. It also covers the order in which a homepage should make its argument. If you would rather have the page read for you, the scorecard does the first pass in a minute. The positioning engagement is for the companies where question one is the hard part.