Why We Built a Solution Finder Instead of a Contact Form

Our contact form had a problem. People would fill in their name, email, and a message that said some variation of "I need a website." No budget. No timeline. No sense of which of our seven services they actually needed. We'd respond, ask five clarifying questions, wait three days for a reply, then ask three more. Half the time, the thread died there. We were losing leads not because we lacked interest, but because we were asking people to do the hardest part of the conversation first: figure out what they needed before they had any guidance.

So we killed the contact form as our primary call to action and replaced it with a Solution Finder quiz. It's a seven-step interactive assessment that asks smart questions, narrows the problem, and generates a personalized recommendation. The difference in how prospects engage with us has been significant. And the thinking behind it applies to far more than just our site.

The Open-Ended Form Problem

A contact form with a big empty textarea is the digital equivalent of walking into a doctor's office and being asked "So, what's wrong?" before anyone takes your vitals. Most people freeze. They either write too little ("I need help with my website") or too much (a 400-word essay about their business history that buries the actual question three paragraphs in).

This isn't a new observation. Baymard Institute research shows that 26% of users abandon checkout flows because the process is too long or complex, and that's in e-commerce, where users already know what they want. Imagine the abandonment rate when visitors don't even know what service to ask about. The Interaction Design Foundation calls this progressive disclosure: only present information and options when they become relevant, not all at once.

Open-ended forms put the cognitive burden on the visitor. They have to self-diagnose their problem, self-select the right service, estimate their own budget, and articulate all of that in a textarea with a blinking cursor. That's a lot to ask from someone who found you through a Google search 90 seconds ago.

The contact form rewards people who already know exactly what they need. The Solution Finder helps everyone else figure it out.

And "everyone else" is most of your audience. The prospects who know they need a WordPress migration with ACF integration and a three-month timeline? They're going to contact you regardless of your form design. The ones who think their site is slow but aren't sure if they need a redesign, a new host, or just some performance work? Those are the people you lose to a blank textarea.

What the Solution Finder Actually Does

Our quiz walks visitors through seven questions in a specific order. The sequence matters because each question builds context for the next.

It starts broad: "What brings you here today?" with options like new website, redesign, fix an existing site, email marketing, accessibility, AI search, or "I'm not sure." That last option is doing heavy lifting. It gives uncertain visitors permission to engage without pretending they have answers they don't.

Then it gets progressively more specific. Current platform. Biggest frustration. How often they update content. What matters most to them. Timeline. Budget range.

Each question is a single screen with large, tappable options. No typing. No free text until the very end. The whole thing takes about two minutes. We timed it.

The recommendation engine runs entirely client-side in vanilla JavaScript. No API calls, no server round-trips, no loading spinners. When you click through the last question, the logic evaluates your answer combinations against 12+ possible outcomes and returns a specific recommendation — something like "WordPress Redesign & Modernization" or "AI Search Optimization Package" — with a short explanation of why that's the right fit and a set of relevant service tags.

The engine uses weighted matching rather than rigid decision trees. If your goal is "fix my existing site" and your frustration is "it's painfully slow" and your priority is "speed," you get the Performance & Speed Optimization recommendation. But if that same person also mentioned WordPress as their platform and said their CMS is painful to use, the engine can weigh those signals and surface the CMS & Admin Experience Overhaul instead. It's not a simple if-then lookup. The combinations are layered.

This matters because we've all taken quizzes that clearly have five predetermined outcomes and the questions are just window dressing. Ours actually branches differently based on real answer patterns. Someone selecting "accessibility" as their goal triggers different downstream logic than someone who arrives through "email marketing," even if their budgets and timelines are identical.

Why This Converts Better Than a Form

Interactive assessments convert at rates that make traditional forms look broken. Research from Interact and similar platforms puts quiz completion rates around 40%, compared to 2-5% for standard contact forms. We don't have those exact numbers for our own site. We're a small consultancy, not a SaaS company with 50,000 monthly visitors. But the qualitative difference is obvious.

Before the quiz, a typical inbound message was vague. After the quiz, every submission arrives with structured data: their goal, platform, pain points, priorities, timeline, budget range, and our specific recommendation. That first reply email we send isn't "Tell us more about your project." It's "Based on your answers, here's what we'd suggest and here's roughly what that looks like." The conversation starts at step three instead of step zero.

Three psychological principles are doing the work here.

Progressive commitment. Each question is small and easy. You're not committing to a project by answering "What brings you here today?" But by the time you've answered seven questions and seen a recommendation that feels accurate, you've invested two minutes of thought. That micro-investment makes you more likely to take the next step. Behavioural psychologists call this the consistency principle — people are more likely to follow through on a larger action after they've already taken smaller related ones.

Reduced choice paralysis. A blank textarea is infinite choice. A set of four to seven options on each screen is manageable. The Iyengar and Lepper jam study from 2000 demonstrated this clearly: 24 choices led to 3% conversion, while six choices led to 30%. Our quiz never shows more than seven options on a single screen. Usually it's four or five.

Personalized outcome. The recommendation at the end isn't generic. Seeing "AI Search Optimization Package" with a description that references your specific pain point creates a moment of recognition. Someone understood what you need. That trust transfer happens before we've spoken a single word.

We wrote about conversion design principles in a previous article, and the Solution Finder applies nearly every one of those ideas. Clear value proposition, reduced friction, immediate feedback, single focused action per step. The quiz is a conversion funnel disguised as a helpful tool.

Building the Recommendation Engine

We considered third-party quiz tools. Typeform, Outgrow, Interact. There's no shortage of platforms that will let you build a multi-step quiz with conditional logic. We chose to build ours from scratch for three reasons.

First, control. Third-party tools impose their own styling, load their own JavaScript bundles, and often redirect users to a hosted page. We wanted the quiz to feel native to the site. Same fonts, same colours, same interaction patterns. The modal opens on top of our page, not inside an iframe from someone else's domain.

Second, performance. Our quiz is vanilla JavaScript. No framework, no dependencies, no external requests. It loads instantly because it's already part of the page. A third-party embed would add hundreds of kilobytes and create a noticeable delay, which matters when the quiz is our primary call to action across the entire site. It's in the nav, the hero, a post-services banner, a pre-insights callout, the contact section, and a floating button. That's seven touchpoints. The quiz has to be fast at every one of them.

Third, privacy. Client-side means no visitor data goes anywhere until they choose to submit their results. No tracking pixels from a SaaS provider. No analytics cookies from a third-party domain. The quiz runs in the browser, calculates a result, and shows it. Only when you fill in the contact form at the end do your answers leave the device. For a company that talks about accessible, privacy-respecting design, that consistency matters.

The actual JavaScript is surprisingly compact. Each question is an object with an ID, the question text, and an array of options. Each option stores a value and optional tags. The recommendation engine maps answer combinations to outcomes using a scoring function. It checks for specific high-priority matches first (someone who selected "email" as their goal always gets the Email Marketing Overhaul, regardless of other answers), then falls back to weighted pattern matching for the less obvious combinations.

Why We Built Custom Instead of Using a Platform

Native styling — Same fonts, colours, and spacing as the rest of the site. No iframe, no embed.
Zero dependencies — Vanilla JS, no external requests, loads instantly from any of seven trigger points.
Client-side only — No visitor data transmitted until the user explicitly submits. No third-party tracking.
Real branching logic — 12+ distinct outcomes based on weighted answer combinations, not a simple decision tree.
Accessible by default — role="dialog", aria-modal="true", arrow key navigation, keyboard focus management, escape-to-close.
Two-minute completion — Seven screens, no typing required until the final contact form.

What We've Learned From the Results

The quiz has been live across the site for a while now, and a few patterns stand out.

The "I'm not sure" option on the first question gets selected more than we expected. Around a quarter of people who start the quiz don't have a clear goal in mind. That's exactly the audience a contact form fails. These are people who know something is wrong with their web presence but can't articulate what. The quiz gives them a structure for thinking about it, and the recommendation gives them language they can bring to the conversation.

Budget transparency changed the whole conversation. We included a budget question with ranges: under $5,000, $5,000-$15,000, $15,000-$30,000, $30,000+, and "not sure yet." Before the quiz, budget conversations happened two or three emails into the thread and were often awkward. Now they're handled before the first hello. Nobody feels put on the spot because they're selecting a range from options, not typing a number into a field. We've written about why asking the right questions before hiring a web developer matters, and the quiz essentially automates that pre-qualification on both sides.

The recommendation accuracy is high. We track whether the actual project we end up doing matches the quiz recommendation, and it lines up about 80% of the time. The other 20% usually isn't wrong — it's adjacent. Someone gets recommended a WordPress Redesign but through conversation we realize a platform migration makes more sense. The quiz got them to the right neighbourhood.

One thing that surprised us: the quiz works well as a tool we can send to existing prospects, not just new visitors. When someone emails us cold with a vague request, we'll sometimes reply with "Take our two-minute assessment and we can use your results as a starting point." It reframes the conversation from "convince us to hire you" to "let us help you understand what you need." That shift in tone is worth more than any conversion rate statistic.

We're not going to pretend a quiz replaces a contact form in every situation, though. If your business sells one product or one service, a guided assessment adds unnecessary complexity. If your prospects are highly technical and know exactly what they want, a straightforward form with specific fields is faster and more respectful of their time.

The Solution Finder model works best when your prospects face a real decision matrix. We offer seven distinct services that can overlap and combine. A business owner who thinks they need a new website might actually need an accessibility audit and some SEO work on their existing one. The quiz surfaces those possibilities in a way a blank textarea never could.

It also only works if the recommendations are honest. We could have built the engine to always recommend the most expensive service package. We didn't. Some prospects genuinely need a Discovery & Strategy Session — a $2,000-$5,000 engagement — and not a $25,000 custom build. The quiz tells them that. Trust is a better long-term strategy than upselling.

If you're considering building something similar, start with the outcomes. List every recommendation you could make, then work backward to the questions that distinguish between them. If two outcomes have identical question paths, one of them probably isn't distinct enough to exist. Our 12+ outcomes each have a clear set of signals that trigger them, and that clarity took more iteration than the JavaScript.

The contact form still exists on our site. It's the secondary option below the quiz on every page. Some people prefer to just write to us, and that's fine. But the quiz is the primary path because it produces better conversations, faster timelines, and prospects who arrive already understanding what we do. That's a better experience for everyone.