SMF·OUTGROW
Can a prompt replace Outgrow?
Forms & surveys — forms, surveys and lead capture
Exhibit tracking slip
Verdict
A calculator is inputs, a formula and a result page, and the honest engineering problem is evaluating a user-authored formula without handing anyone a code-execution hole in your server. A small expression parser solves that properly in a day. The rest of Outgrow is the packaging: templates that look designed, and the integrations that push a completed calculation into whatever the marketing team uses next.
Exhibit A — The prompt
Received on31.07.2026Build a calculator builder: user-defined inputs, a safe formula evaluator, and a result page with an optional lead gate.
Inputs: number with a min, max, step and unit; slider; select with a numeric value per option; toggle; and a text field for display only. Each input has a variable name used in formulas, a label, help text and a default.
The formula evaluator, which is the part to get right. Implement a small expression parser and evaluator — not a language runtime, and never any form of eval on user input. Support:
- arithmetic operators, parentheses, and comparison operators
- a conditional expression (if condition then value else value)
- a fixed set of functions: min, max, round, floor, ceil, abs, sqrt, pow
- references to input variables and to previously computed intermediate variables
Parse once at save time, report errors with a position in the expression, and reject anything referencing an undefined variable. Evaluate with a step limit and no access to anything outside the declared variables. Write tests for division by zero, missing inputs, and a circular reference between intermediates.
Intermediates: a calculator can define named intermediate values computed from inputs, and the result page can show any of them. This is what makes a real calculator legible — showing only the final number leaves the visitor unable to see how it was reached.
Result page: the headline number formatted with a unit and decimals, the intermediates as a breakdown, and outcome text chosen from author-defined ranges over the result. Optionally a simple bar or gauge.
Lead gate: off by default. When enabled, the visitor sees the headline number and the breakdown is revealed after submitting an email address. Store the email with the full input set and result, plus a consent timestamp and the wording they consented to. Never gate the whole result behind the form — a calculator that shows nothing until you pay with an email is the pattern that makes people distrust the whole category, and the interface should say so when the option is enabled.
Delivery: a standalone page at a stable slug, plus an embed script rendering into a shadow root, sized to content, under 25 KB.
Data: every completed calculation stored with its inputs and result, exportable as CSV. That dataset — what people actually entered — is more valuable than the leads, and most calculator products bury it.
Accessibility: real labelled form controls, keyboard operable, result announced to assistive technology on change.
Out of scope: a template library, CRM or marketing-automation integrations, A/B testing, and any hosted multi-tenant deployment.Opening prefills the prompt — press enter to run it.
Exhibit B — What you lose
- B.1 the template library that makes a calculator look professionally designed
- B.2 the CRM and marketing-automation integrations
- B.3 the analytics on where in the calculator people abandon
- B.4 hosted delivery and the uptime that comes with it
Prior art
Exhibit C — Why people still pay: integrations, deliverability, and workflow depth
Because a calculator is a marketing asset, and marketing teams buy the template and the integration rather than the arithmetic. The formula is the easy part of the job.
Questions
Can I import my Outgrow calculators?
No structured export exists, so each calculator is described again — inputs, formula, result ranges. For a handful that is an afternoon, and the formulas usually get simpler in the retelling. Past submissions export as CSV for your records.
Why write an expression parser instead of using eval?
Because the formula is authored in an interface and evaluated on a server, which makes eval a direct path from a form field to running code. A small parser with a fixed function list and a step limit takes a day and removes the entire class of problem.
What does it cost to run?
A small VPS with a database, five to ten dollars a month, regardless of how many calculators or completions. Against Outgrow's Freelancer plan this is a clear saving if you already run a server and roughly a wash if you do not.
What is the one thing that does not survive the rebuild?
Looking designed. Outgrow's templates are why a calculator embedded on a landing page reads as a professional tool rather than a form. A functional calculator you styled yourself works perfectly and converts worse, and that difference is the whole point of the product.
Related tools
Receipt