Summarize with AI
A SaaS website for multiple audiences works best when the homepage picks one of them. Give the hero to the audience that arrives cold, which is usually the person who will use the product, and give the other audience, usually the person who signs the contract, a page of its own that the nav and the hero both point to. A hero split in two, with two screenshots and two buttons of equal weight, serves neither reader.
This matters most when the user and the buyer are different people. You know both of them. You have sat on calls with each one and you can tell them apart in a sentence. A stranger on your homepage is only ever one of them, and a page that talks to both at once hands that stranger the sorting you already did in your head.
The hero that says Better writing, for every classroom
Picture an AI tool that helps middle and high school English teachers give feedback on essays. This is an example, not a client. A teacher connects Google Classroom, picks an assignment and a rubric, and the tool drafts comments on every essay in the class set. On paragraph two of one essay, a comment reads: Your claim is clear, but the quote after it does not support it. Which line from chapter 4 would? Beside it, the rubric row Evidence shows 2 of 4. The teacher edits the comment, clicks Approve and moves on. A stack of 28 essays that used to eat a Sunday now fits into an evening.
The people who pay are somebody else. A school district buys it for every English teacher in its middle and high schools. The person who signs is a director of curriculum. Before she signs, she needs a student data privacy agreement, rostering through Clever or ClassLink, a dashboard of writing scores by school, and a price she can put in next year's budget.
The homepage tries to speak to both of them. The headline reads Better writing, for every classroom. Under it sits a pill toggle with two options, I'm a teacher and I'm an administrator, and neither is selected. The left half of the hero shows the comment panel, shrunk until the comments are grey lines. The right half shows a district dashboard with three tiles, Average rubric score, Essays reviewed and Schools active, just as small. Two buttons of the same size sit underneath: Try it free and Book a district demo. On a phone the halves stack, so the first picture a teacher sees under the headline is a chart of schools.
Eleven district demos, seven of them teachers
Now watch a normal month from the founder's side. Most homepage visits come from Google searches like essay feedback tool and from links shared in teacher groups. The sign up form asks for a role, and of last month's 412 sign ups, 389 picked Teacher.
Then he opens his calendar. Book a district demo brought in eleven calls. Seven were single teachers who picked the bigger sounding button and wanted to know how to import a class from Google Classroom. He took each one as a half hour video call and sent the help article afterwards.
The district directors he wants did not come from the homepage first. Two arrived through a link a teacher forwarded, with a note along the lines of could we get this for the whole department. One stopped at his booth at a conference. When they did open the homepage, they clicked I'm an administrator and got the same hero with the dashboard a little larger. Their questions about student data and rostering were answered nowhere above the footer. He has rewritten the headline three times this school year, and each version tried harder to fit both readers in.
What a hero split in two costs
The teacher side pays in activation. Measured on SaaS and product-led companies, activation of 20 to 40 percent is normal and 40 to 60 is best in class, and a ten point improvement typically drives a 15 to 25 percent increase in free-to-paid conversion. A teacher who has to work out whether the page is for her, then read a chart of schools before she sees a single comment, starts her trial already less sure.
The buyer side pays in the founder's week. Seven wrong calls a month, with the prep and the follow-up, is close to a full working day spent on support questions, taken by the one person who can sell to a district. Across all sectors, the median gap between seed and Series A is around 616 days, and roughly 70 percent of seed companies never raise one. In this example, district contracts are how the company gets to the seed bar for SaaS of roughly 300 to 500 thousand in ARR, and the page those buyers need does not exist yet.
Why the toggle is the wrong default
The usual advice for a SaaS website with more than one audience is to segment on the homepage. Add a toggle, a pair of tabs, a row of I am a cards, or a headline broad enough that everybody fits. It sounds fair to both readers. It is the wrong default.
A toggle asks a stranger to sort herself before she knows what the product is. A broad headline fits both readers because it says nothing specific to either one. And a split hero halves the room for the one thing a cold visitor needs, which is to see the product doing its job at a size she can read.
The buyer does not need the hero anyway. A director of curriculum is not deciding in four seconds whether the product deserves a look. A teacher already decided that for her. She arrives with a list of questions, and a hero has no room for a single one of them. She needs a page with room.
There is one exception. When both audiences need the same first claim, one story can carry them. Mi-VAD, a heart pump the size of a coin, had two rooms to win, investors and clinicians, and its site was built as one story for both. The test is the first proof each reader needs. When one needs a screenshot and the other needs a data agreement, they need different pages.
Find out who arrives cold before you change a word
Most founders guess, and they guess the audience they talk to most. Check instead, in one afternoon:
- In analytics, look at visits that land on the homepage from search, social and referral links. Leave out direct visits and searches for your name, since those people already know you. Read the top referrers and the search terms.
- Read the role or job title field on sign ups and demo requests from the last three months. Count each answer. If the form has no such field, add one this week.
- For your last five paying customers, find the first person at that company who touched the site. Your CRM or the sign up date on the account usually tells you.
- Watch five session recordings of strangers on the homepage.
The decision rule: whichever audience makes up most first visits from strangers owns the hero. In a product-led tool like the essay example, that is almost always the user. In a tool sold through outbound email to a senior buyer, it can be the buyer, and then the user is the one who gets a page of her own. If the second audience is really a buyer you are moving toward rather than one you serve alongside the first, that is a different job, and the pages that change first when you reposition your ICP are not the same ones.
What the For districts page carries
Give the second audience a page with a plain address, like /districts, and a nav link that names them, For districts, placed just before Sign in. Then fill it with the answers she brings, in the order she brings them:
- A first line that names her and her job: Essay feedback for every English teacher in your district, with student data that stays yours.
- Proof from her world. Name the districts already on a contract, and let her type her district's name to see how many of her own teachers already use the free plan.
- The student data privacy agreement as a download, the states where it is already signed, and one plain sentence on whether essays ever train a model.
- Rostering: the Clever and ClassLink logos, with one line each on what syncs and who sets it up.
- The district dashboard at full width, the same screen that was squeezed into half a hero.
- What a pilot looks like: one school, one term, who runs the training, and what she will have to show her board at the end.
- How the price is counted, per school per year, even if the numbers live on a call.
- One button: Book a district call.
Our guide to the enterprise page covers the same job for a corporate buyer and most of its rules carry over.
The hero keeps one reader and one quiet line
Back on the homepage, the hero now belongs to the teacher. The headline reads Feedback on a class set of essays in one evening, in your own words. The comment panel fills the visual, one essay at a time, readable on a phone. There is one button, Try it on your next essay set. The toggle is gone.
Under the button, in smaller type, sits one line for the other reader: Buying for a school or district? See how districts roll it out. That line does the routing the toggle tried to do, and only the readers it names ever click it.
Put the same door inside the product, where the forwarding actually happens. After a teacher approves her third class set, show a small card: Want this for your whole department? Send this page to your curriculum lead. In this example that is how most district buyers arrive, so the link belongs where the teacher already is. Then change the Book a district demo form to ask for the district name, so a teacher who still lands there gets sent to the help center instead of your calendar.
If you would rather hand the work to a studio, our ranking of website design agencies for B2B AI SaaS companies compares the ones that take on this kind of rebuild.
Two audiences are hard to sort from the inside, because you already know which reader is which and the page never has to tell you. For AI founders in the US, UK and Europe, Studio Maydit does website design and continues into the product behind it, in Framer, Webflow or custom code. A fixed-scope project takes three to four weeks. If your hero is trying to serve two readers at once, book a call with us at cal.com/maydit/45min.
