What to bring to a first conversation so the estimate you get back is worth reading — and what you can safely leave until later.

A good brief is not a specification. It is the smallest amount of information that lets someone tell you honestly whether they can help.

What to bring

You do not need wireframes, and you certainly do not need a page count. You need to be able to answer four questions.

1. Who is the site for, and what do they currently believe?
2. What has to change about that belief?
3. What happens after someone is convinced — a form, a call, a purchase?
4. What has already been tried?

The fourth is the one most often left out, and the one that saves the most time. An agency that knows you already tried a rebrand two years ago and it did not move enquiries will ask different questions than one that does not — and will stop you from paying to relearn the same lesson.

Why "why now" matters more than it sounds like it should

Every business could theoretically use a better website at any point in its history. The businesses that get real value from the project are the ones where something specific changed: a new product line, a competitor that just launched something sharper, a founder who noticed the site has not been touched since a previous hire built it four years ago and left. "Why now" tells an agency what the site actually has to prove, this year, for this reason — as opposed to "make it better," which is true of almost every website on the internet and therefore tells nobody anything useful about what to prioritise.

What you can leave out

Colour preferences. Competitor sites you admire. A sitemap. All of these are useful eventually, and all of them narrow the conversation before it has usefully widened. A sitemap in particular tends to lock in a structure before anyone has agreed what each page is supposed to prove — and once a client has handed over a sitemap, changing it later feels like scope creep even when it is exactly the correction the project needs.

Competitor references are worth mentioning but worth holding loosely. "We like how clean this site is" is useful signal about taste. It is not useful signal about whether that competitor's structure will work for your business, your audience, or the argument you actually need your site to make — two companies in the same category can need entirely different sites if one sells on price and the other sells on expertise.

Budget

Say the number. An agency that changes its recommendation based on your budget was not going to give you an honest recommendation anyway, and one that does not know your budget cannot tell you what is realistic inside it. A range is enough — "somewhere between X and Y" lets an agency tell you immediately whether the project you are describing fits inside it, or whether the scope needs to shrink, or whether this is not the right agency for the number.

Withholding the budget "to see what they come back with" is a common instinct and a costly one. It produces a proposal built around guessing your ceiling rather than around your actual priorities, and guessed proposals are either padded for safety or optimistic to win the work — neither of which is the estimate you actually wanted.

What a first conversation should cover, in order

A useful first call moves through a predictable sequence: the commercial goal first, then the audience, then what exists already — content, brand assets, a previous site worth salvaging pieces of — and only at the end does layout or design style come up at all. Agencies that open by asking about colour palettes and font preferences are, whether they realise it or not, treating the project as a decoration job rather than an argument that needs building. That is a reasonable signal to notice early, before a proposal is written around the wrong set of questions.

What you should get back

A scope, a price, and a clear account of what is excluded. If the exclusions section is empty, the estimate has not been thought about — every real scope excludes something, whether that is copywriting, a specific integration, or ongoing maintenance after launch, and naming those exclusions explicitly is what prevents them from becoming disputes in month two.

The scope should also name what happens if an assumption turns out to be wrong. "We assumed your product catalogue was under two hundred items; if it's larger, the information architecture changes and so does the price" is the kind of sentence that protects both sides, because it turns a possible future disagreement into something both parties already agreed to before it happened.

A brief that works, in miniature

The clearest briefs we get read something like this: "We sell industrial parts to procurement managers who currently find us through a distributor catalogue, not our own site. We want direct enquiries instead, because the distributor's margin is eating our growth. We tried a rebrand two years ago that didn't move this number, because it changed how we looked without changing what a procurement manager could actually do on the site. Budget is X. We need this live before our trade show in October." That paragraph contains everything from the four questions above, in four sentences, and an agency that receives it can give you an honest answer inside a day.

What happens when the brief is thin

A thin brief does not stop a project from happening. It just moves the missing questions later, to a point where answering them is more expensive. An agency working from "we need a new website, something modern" will make assumptions to fill every gap you did not — about audience, about tone, about what "modern" is supposed to mean for your specific business — and every one of those assumptions is a small bet that may or may not match what you actually needed. Discovering a mismatched assumption in week six of a build costs a revision cycle. Naming it in the brief costs one sentence.

This is not a case for over-preparing before the first call. Most of what a brief needs is already in your head — you know why you are doing this now, who the site needs to convince, and what you have already tried that did not work. The work is writing those three things down before the conversation, rather than discovering them out loud during it, because a conversation that starts from real answers moves considerably faster than one that starts from "well, what do you think we need?"

Local specifics worth mentioning

If the project has anything Bucharest- or Romania-specific about it — a bilingual audience, a regulatory requirement particular to the market, a competitor set that only makes sense locally — say so in the brief rather than assuming it is obvious from your address. An agency that works across markets will structure the project differently for a business that needs to argue in two languages to two audiences with different expectations than for one serving a single-language market, and that difference is cheaper to design for from the start than to retrofit once the information architecture already assumes one language is primary.

In short

Bring the commercial reason, the audience, what has already been tried, and a real number. Leave the colours and the sitemap for later — they are easier to get right once the argument underneath them is settled, and considerably harder to get right before it is.