Segmenting with a quiz isn't about "collecting more customer data." It's about splitting incoming leads so you handle different people differently: call some today, send others a set of options, and politely say no to the rest.
The value isn't in the split itself but in what happens next. If every lead still lands in the same pile with the same salesperson, you don't need segmentation: you asked those questions for nothing.
Which criteria to segment by
Useful criteria are few, and each one has to change something about how you handle the lead.
- Readiness to buy. The timeline question, "when do you plan to start": today, this month, just looking. This is the most useful criterion: it tells you whom to call first.
- Project size. Square footage, volume, quantity. It separates small inquiries from large ones and sometimes retail from wholesale.
- Type of need. The service line. It determines which salesperson or team the lead goes to.
- Location. Whether you serve that area or not. The cheapest way to avoid wasting time.
- Budget. Useful, but handle with care: ask in ranges, near the end, with a "not sure yet" option.
A criterion that changes nothing — neither response speed nor who tells the customer what — isn't segmentation. It's an extra question, and it costs you a share of your leads.
How to build it inside the quiz
There are two levels, and it's best to start with the first.
Level 1: qualifying questions
The quiz is the same for everyone, but the answers arrive with the lead, and you use them to route leads by hand or with rules in your CRM. For most businesses this is enough: one timeline question and one size question already give you clear prioritization.
Level 2: branching
The first question splits people by service line, and then each person sees only their own steps. Use it when the lines are truly different: asking someone who came for one thing about something else is a sure way to lose them. How to set it up is in the article on quiz branching.
An important side effect: branching keeps the quiz short for each individual person, even if there are many questions in total.
What to do with the segments afterward
This is where the payoff that started it all shows up.
- Call priority. Someone planning to start this week should get a call before someone who's "just looking." The conversion difference here is bigger than any quiz tweak.
- Routing between salespeople. Leads from different lines go to different people: when you send them to your CRM, you set this up from the start. For example, when sending leads to Kommo (formerly amoCRM), you can set the pipeline and the owner.
- A different call script. The salesperson sees the answers before calling and starts with what matters, not with a questionnaire.
- Different emails. If you collect email, segments let you send something relevant instead of one blast for everyone.
- An honest "no." If you're not a fit for someone, tell them right away: that way they don't sit in your funnel skewing your stats.
Segmentation and lead quality
A side effect, and often the main one: qualifying questions also act as a filter. People you can't serve because of location or volume often drop out on their own when they see the question. On paper, quiz conversion goes down; in practice, you stop paying your salespeople for conversations that go nowhere.
That's why judging segmentation by quiz conversion is a mistake. Judge it by the share of leads that turned into sales. More on this in the article why a quiz brings lots of leads but few sales.
Common mistakes
- Segmenting "for later." You added questions, but there's no process behind them: every lead is still handled the same way. Pure conversion loss.
- Too many segments. If there are more segments than salespeople, the setup won't work.
- Questions for your internal analytics. "How did you hear about us?" serves you, not the customer, and costs you part of your audience. Get the source from UTM tags instead.
- Budget as the first question. The most common way to lose people before they see any value.
- Segments without priority. Splitting the flow and handling it all in the same order is the same as not splitting it.
FAQ
How many questions do I need to segment?
Usually two or three qualifying questions within the usual 4 to 6 questions are enough. You don't need extra questions "for segmentation": the same size and timeline questions do both jobs.
Do I need a separate quiz for each segment?
Almost never. Branching is enough: the first question splits people by line, and then each person sees only their own steps.
Should I ask about budget?
Yes, if it affects your offer, but in ranges, near the end, with a "not sure yet" option. Placed at the start, that question collapses the funnel.
How do quiz answers get into my CRM?
Through the integration: the answers are sent along with the contact, and you set up routing based on them. One example: sending leads to Kommo.
Conversion dropped after I added qualifying questions. Is that bad?
Not necessarily. If the people you can't serve dropped out and the share reaching a sale went up, you're winning. Look at sales, not quiz conversion.
