How to Read a Website Redesign Proposal Before You Sign It

Almost every guide to website proposals is written for the person sending it — how a designer should structure a proposal to win the work. This is the other side of the table: how to read one before you sign, so that the ambiguity in it becomes your question today instead of your delay, change order, or ownership fight three months from now.

There's a single test that does most of the work. Hand the proposal to a capable person who wasn't on any of the sales calls and ask them to explain what both sides are about to do. If they can — the current problem, who the site is for, what's being built, who's responsible for what, what it costs now and later, and how you'll know it's finished — the proposal is doing its job. If they keep having to say "I think this is what they mean," the gaps are real, and a polished mockup won't fill them.

A quick boundary before the details: this is general business education, not legal advice. A proposal might be folded into a contract, sit beside a statement of work, or contain binding terms on its own — the label on the document matters far less than what the whole agreement says. Have qualified counsel review the agreement when the cost or risk warrants it, and have qualified counsel review ownership and exit terms specifically.

---

Website Proposal Scorecard

Use: Buyer worksheet for scoring and comparing website proposals. Print it or save it as a PDF with the button above and score each proposal on paper. General business education, not legal advice.

A proposal is useful when a capable person who missed the sales calls can explain the problem, scope, responsibilities, evidence, price, ownership, launch decision, ongoing operation, and exit without guessing.

This scorecard turns those questions into a repeatable comparison. It does not make unlike proposals identical, and a high score does not repair a legal, financial, security, or ownership term that requires qualified review.

How scoring works

Score each category from 0 to 5, then multiply by the category weight.

• 0 — absent: the proposal does not address it.
• 1 — implied: reassuring words, but no accountable detail.
• 2 — partial: some useful detail with important gaps.
• 3 — workable: responsibilities and deliverables are mostly clear; limited clarification remains.
• 4 — strong: clear scope, owner, evidence, boundary, and change process.
• 5 — inspectable: a third party could execute or verify the term from the document set without relying on sales-call memory.

Weighted points equal score ÷ 5 × weight. The total possible score is 100.

Automatic stop-and-clarify conditions

Do not rely on the total score when any of these remain unresolved:

• the complete agreement set or order of precedence is unclear;
• the business cannot identify who controls the domain, DNS, site, content, data, or critical accounts;
• lead capture is promised but no durable source record or acceptance test is defined;
• a provider guarantees rankings, AI recommendations, lead volume, or sales outcomes it does not control;
• recurring fees, cancellation effects, exports, or operated components that stop are unclear;
• the price depends on undefined “standard” pages, integrations, revisions, or client responsibilities;
• the proposal incorporates external terms that the buyer has not received; or
• a material legal, privacy, accessibility, licensing, or regulatory issue needs qualified review.

The 100-point scorecard

Evidence requests by category

Diagnosis

Ask for:

• a short current-state inventory;
• examples of verified problems versus assumptions;
• the least expensive viable alternative; and
• the condition that would make a rebuild unnecessary.

Outcomes and boundaries

Ask the provider to finish these sentences:

• “The website will…”
• “We will advise on…”
• “The client will operate…”
• “We will measure…”
• “We do not promise…”

Scope and deliverables

Require nouns and quantities where they matter. “Website,” “SEO,” “copy,” “analytics,” and “integration” are categories, not deliverables.

Useful evidence includes a page inventory, component list, content responsibility table, redirect inventory, form list, profile list, analytics event map, and launch checklist.

Responsibility matrix

For every dependency, name:

• who supplies it;
• who decides;
• who implements;
• who verifies;
• who operates it after launch; and
• what happens when it is late or unavailable.

Evidence and acceptance

Ask what you will be shown before approval:

• mobile and keyboard contact-path tests;
• a durable synthetic inquiry record;
• notification attempt and technical receipt evidence;
• canonical, sitemap, redirect, and indexability checks;
• accessibility checks and known limitations;
• analytics event definitions; and
• the ownership/access handoff packet.

Ownership, access, and data

Ask for an asset schedule with owner, administrator, billing party, export method, license, recovery path, and exit disposition. Never put passwords or secret values in the proposal.

Price and ongoing cost

Separate:

1. one-time build work;
2. third-party and pass-through subscriptions;
3. ongoing operation and support; and
4. future changes outside the included allowance.

Timeline and change control

A date without dependencies is theater. Ask which decisions control the path, what review window is assumed, how silence is handled, and what changes the price or launch date.

Exit and continuity

Ask what the business receives, what the provider transfers, what remains licensed, what operated components stop, how data is exported, and what continued operation would require elsewhere.

Reading the total

• 85–100: unusually clear. Resolve every stop condition and remaining category-specific question before signing.
• 70–84: potentially workable. The gaps should become written clarifications or revisions.
• 50–69: too much depends on sales-call memory or optimistic interpretation.
• Below 50: the buyer is being asked to approve ambiguity.

These ranges are decision aids, not performance data or legal conclusions. A lower-cost proposal can be the right choice when its narrower responsibility is explicit. A higher-cost proposal can still be poor when it sells confidence without inspectable scope.

Comparison record

If you would rather fill that record in on screen, the proposal comparison tool covers the same ground for two or three proposals and writes the clarification questions for you to send. It runs entirely in your browser, keeps no score, and names no winner — the scoring above stays here, where you weigh it yourself.