The Estimate Engine

Storyboard, pricing framework, best practices, and pilot proof of concept for an enterprise build calculator.

1 · What This Is & Why

Large corporations don't buy from a "contact us" form. Their evaluators want to self-serve a ballpark number before they'll put a vendor in front of procurement. The Estimate Engine gives them that number in under two minutes — while quietly demonstrating the exact kind of interactive tooling Vivere builds.

The calculator prices six areas — five build modules plus an organization profile:

2 · Storyboard — The Six Frames

The pilot follows this exact user journey. Each frame below sketches the screen state and the emotional beat we're aiming for.

3 · The Pricing Framework

Everything the calculator shows traces back to one plain-object config — PRICING — at the top of the pilot's script. The interface never hardcodes a dollar figure. That separation is the framework:

CONFIG   (PRICING object — rates, multipliers, tiers)
   ↓
RULES    (compute(state) — pure function, no DOM)
   ↓
STATE    (user's slider/toggle selections, persisted to localStorage)
   ↓
RENDER   (module costs, range band, breakdown bars, print sheet)

The formula

moduleSubtotal = Σ module costs           // each: units × rate × tier multiplier
orgAdders      = SSO + compliance + WCAG + (locales × rate)
sizeOverhead   = moduleSubtotal × (sizeMultiplier − 1)   // mid 1.0 · large 1.2 · global 1.4
oneTime        = baseEngagement + moduleSubtotal + sizeOverhead + orgAdders
shownRange     = oneTime × (1 ± contingency)             // ±15%
timelineWeeks  = max(2, 4 × (oneTime ÷ $20k)^0.55)       // sublinear — budget adds crew, not calendar
monthly        = supportTier rate                        // separate line, never blended in

Design decisions worth defending

• Pure compute function. compute(state) touches no DOM. It can be unit-tested, moved server-side, or reused in a proposal generator without change.
• Size multiplier applies to modules only — not to compliance adders (those are already flat enterprise prices) and not to the base engagement. No double-charging.
• Support is monthly, shown separately. Blending a retainer into a one-time figure is how estimates lose credibility in front of a CFO.
• Timeline derives from the same number — sublinearly. Doubling the budget doesn't double the calendar; it adds crew. The power-curve (exponent 0.55) keeps a $1.5M program at months, not years.

4 · Best Practices

Distilled from what works in enterprise pricing tools generally, and from Vivere's own client-report experience:

1. Ranges, never point prices. A single number reads as a quote and gets negotiated. A ±15% band reads as an honest planning figure and starts a conversation.
2. Show where the money goes. The breakdown bars do more persuasion than the total. Evaluators need to defend the spend internally; give them the ammunition.
3. Price enterprise overhead openly. SSO, compliance, accessibility, and localization appear as visible line items. Burying them in a multiplier is where trust dies in procurement.
4. No gate before the number. The estimate is free and instant. The detailed proposal is what sits behind discovery — consistent with our gated-report standard for execution detail.
5. The exit artifact is a document. Enterprise decisions travel by PDF, not by URL. The print sheet is a first-class feature, not an afterthought.
6. Config over code. Anyone adjusting rates edits one object. If a rate change requires touching markup, the framework has failed.
7. Persist the configuration. Evaluators return, tweak, and re-print. localStorage in the pilot; a shareable config URL in v2.
8. Keyboard and screen-reader clean. Real range inputs, aria-pressed segmented controls, a polite live region on the total. A tool that claims WCAG passes as a line item had better pass one itself.

5 · Pilot Proof of Concept

The pilot is live in this folder as a single self-contained HTML file — no build step, no backend, deployable to Cloudflare Pages as-is.

What's in the pilot (v0.1)

6 · Rollout Ladder & Next Steps