You sent one brief and got back five completely different proposals. Different formats, different prices, different assumptions about what the project even is. Now you have to pick one. This post is for that exact moment.

Most evaluation advice tells you to compare line items. That doesn't work when no two proposals use the same line items. What actually works is running a short, consistent question round with every agency before you decide. The answers expose gaps that the proposals themselves were designed, intentionally or not, to obscure.

First: Normalize what you're actually comparing

Before you ask anything, read every proposal and flag the sections that are vague or missing entirely. Every proposal has them. Common gaps: who owns the code after launch, what happens if the project runs over scope, whether "design" includes mobile, and what "support" means after the site goes live.

Make a list of those gaps. They become your questions. The goal isn't to catch anyone out. It's to get every agency on record about the same set of specifics, so you're comparing answers instead of marketing language.

The questions that matter most

1. Walk me through what happens in the first two weeks after we sign.

This question separates studios with a real process from studios that figure it out as they go. A good answer includes a kickoff structure, how they gather what they need from you, and what you'll see and approve before any building starts. A vague answer, something like "we'll get aligned on goals," is a flag. You want to know what the first deliverable is and when you'll see it.

2. What's not included in this proposal that a project like ours typically needs?

This is the most important question on the list. A piece on how local newsrooms commission web work made the point that buyers often discover missing scope after contracts are signed, not before. Ask this directly and watch what happens. An honest agency will name real things: content migration, staff training, analytics configuration, third-party integrations. An agency that says "everything is included" is either padding the price or setting you up for change orders.

3. Who will actually do the work?

Some studios sell work and subcontract all of it. Others have in-house teams but staff projects with junior people once the senior lead closes the deal. Neither is automatically wrong, but you deserve to know. Ask for the name and role of the person who will build your site, and ask whether that person is an employee or a contractor. Then ask if that person is available to start when the proposal says the project begins.

4. What do you need from us, and when?

Projects at small media organizations and nonprofits stall because the client side gets busy. A good agency has thought about this and will tell you exactly what they need from you, content, decisions, feedback windows, access to existing systems, and when they need it. If an agency can't answer this question, they haven't planned the project yet. They've only priced it.

5. What happens if this takes longer than scoped?

Every project has moments where something takes longer than expected. The question is who absorbs that. Some agencies charge change orders for anything outside the original scope document. Others build buffer in and hold the line on price unless the scope genuinely changes. Neither model is wrong, but you need to know which one you're buying. Ask for a specific example of how they've handled this with a past client.

After launch: the question most buyers forget

6. What does ongoing support look like, and what does it cost?

A lot of proposals include "30 days of post-launch support" and leave it there. Ask what that actually covers. Bug fixes only? Content updates? Training for your team? What happens on day 31? For a small newsroom or nonprofit, the period right after launch is often when the real questions start. You need to know whether you'll have a partner or a closed ticket at that point.

How to use the answers

Send the same questions to every agency in the same week. Give them a deadline. The speed and specificity of the responses are themselves useful data. An agency that takes ten days to answer a six-question email is showing you something about how they'll communicate once you're a client.

Once you have answers, score each agency on the questions that matter most to your organization. For most small publishers and nonprofits, that's clarity on scope, post-launch support, and who is actually doing the work. Price matters, but it matters less than whether the project will actually land where you need it to.

The proposal that wins shouldn't be the one that looked best on paper. It should be the one that held up when you pushed on it.