Is building a website before pitching just working for free?

Only if you build a full custom site for every prospect. The method works because the first version is a template populated with the business's public information, produced in minutes rather than days. You are investing a small fixed cost to replace an unpaid sales conversation, not delivering a finished custom project for free.

What should the demo site actually contain?

Only public information: the business name, what it does, its services, its location and opening hours, its contact details, and its existing reviews or photos where publicly available. It should look finished on a phone and load fast. It should not contain invented claims, fake testimonials, prices you guessed, or the business's logo used in a way that suggests they approved it.

What do I charge after sending a demo site?

Quote a fixed price for a defined scope rather than an hourly rate. The prospect has already seen roughly what they are getting, which makes a fixed number easy to say yes to. Ambiguity is what kills these deals, so state exactly what is included, what it costs, and how long it takes.

Do I need the client's permission before building a demo?

You do not need permission to build an unlisted demo from public information, but you should keep it unindexed, never present it as their live website, make clear it is an unsolicited proposal you made for them, and take it down promptly if they ask. Be careful with logos and photos, which may be protected, and follow the marketing rules that apply in your country and theirs.

client acquisition

Build the Website Before You Pitch: A Better Way to Get Web Design Clients

DreamDevs Team·Aug 2, 2026·10 min read

Designing a website demo before pitching it to a prospective client

The standard advice for winning web design clients is to send a cold email introducing yourself, mention that you noticed they do not have a website, and ask if they would like to hop on a call. It is polite, it is professional, and it almost never works.

We know because we ran it. Then we changed one thing: instead of asking whether they wanted a website, we built the website and sent them the link. The response rate was not marginally better. It was a different business.

Why the Standard Cold Email Fails

A typical outreach email asks the recipient to do a surprising amount of unpaid work before they can even reply. They have to picture what their website might look like, guess what it would cost, estimate how long it would take, work out how much of their time it would eat, and decide whether a stranger who emailed them can be trusted with any of it.

That is five separate judgements, all requiring imagination, all in a message they did not ask for while doing something else. The rational response is to ignore it, so they do.

The core insight: your prospect is not resisting the website. They are resisting the ambiguity. Remove the ambiguity and you remove most of the objection.

What Changes When You Send a Finished Site

When the email contains a working link to a site that is already built for their business, with their name, their services, their opening hours, and their photos, every one of those judgements collapses into one simple reaction: do I like this or not?

Four things shift at once:

Proof replaces claims. You are not saying you build good websites. They are looking at one, for their business.
The endowment effect kicks in. The moment something has your name on it, letting it disappear feels like a loss. This is a far stronger motivator than the promise of a future gain.
Effort signals sincerity. Everyone knows a template email costs nothing to send. A finished site reads as deliberate, which earns a reply even from people who say no.
The conversation starts further along. Instead of debating whether to have a website, you are discussing whether the contact section should include a booking form. That is a customer conversation, not a sales one.

The Objection: Is This Just Working for Free?

This is the right question, and the answer determines whether the strategy is smart or self destructive.

If you hand craft a bespoke site for every prospect, yes, you will burn weeks and go broke. The method only works when the first version is cheap to produce. That means:

Templates, not custom design. One strong template per vertical, populated with each business's public information.
Public data only. Name, services, location, hours, contact details, public reviews and photos. No research projects.
Minutes, not days. If a demo takes longer than a coffee break, your process is wrong, not your strategy.
The custom work happens after they pay. The demo is the conversation starter. The paid engagement is where their real content, branding and requirements get built in.

Framed properly, you are not giving away a project. You are spending a few minutes of template time to replace a sales call that would have cost you thirty minutes and probably would not have happened.

This is also why the strategy only became practical recently. When a site took days to produce, building one speculatively was irrational. Now that generation costs cents and minutes, the maths inverts. We unpacked that shift in AI website builder agents and what they actually do for agencies.

Who to Target

This play works best on businesses that are visibly trading but visibly underserved online: no website at all, or one that is broken, unresponsive on mobile, or years out of date. Prioritise verticals where one new customer pays for the whole site, which makes the decision easy on their side.

We covered the sourcing and screening process in detail in how to find businesses without a website. Do that part first, because sending beautiful demos to businesses that cannot pay is just an expensive hobby.

What the Demo Site Must and Must Not Do

Get this wrong and the play backfires. The rules:

It must look finished on a phone. Most of them will open it on a phone. A demo that breaks on mobile proves the opposite of your point.
It must load fast and use their real details. Wrong phone number, wrong hours, or a misspelled business name destroys the whole effect instantly.
It must not invent things. No fabricated testimonials, no made up statistics, no awards they never won, no prices you guessed. If it is not publicly true, it does not go on the page.
It must not pretend to be live. Keep it on an unlisted URL, keep it out of search engines, and never imply it is their actual website or that they commissioned it.
Be careful with logos and photos. These are often protected. Using them to mock up a proposal for that business is normally fine, but take the demo down promptly if you are asked to.

Also check the marketing and anti spam rules that apply where you and the recipient are based, identify yourself honestly, and always give people a clear way to opt out. This approach earns replies because it is genuinely useful, and that only holds if you are straight with people.

The Message That Goes With It

Keep it short. The site is doing the persuading, so the email only has to get the link opened. A structure that works:

One line naming the specific thing you noticed, with no criticism of their business
One line saying you built something for them and here is the link
One line on what happens next if they like it, including the price or a clear price range
One line making it painless to say no

Do not attach a PDF, do not send a deck, do not ask for a meeting in the first email. The only goal is a click. Give them the price early: the fear of an unknown quote is a bigger blocker than the number itself usually is.

"Cold emails that ask a question get ignored. Building the site first worked far better, so we automated the tedious half." - DreamDevs

Then Handle the Replies Properly

Expect three kinds of response.

Interested but hesitant is the most common. They like it and want to know about cost, ownership, changes, and who maintains it. Answer plainly with a fixed price and a defined scope. Vagueness here loses deals you had already won.

Interested but wants changes is the best outcome, because they are already treating it as their site. Make one round of changes, then ask for the decision.

Not interested is fine. Thank them, take the demo down, and move on. Some of them come back months later, which does not happen when you have been pushy.

Doing This at Volume

The strategy is simple. The bottleneck is production. Screening each business, checking whether their site loads and works on mobile, picking a template, populating it with real details, publishing it, and sending the message is perhaps twenty minutes of unskilled work per prospect. At fifty prospects, that is your entire week.

That is exactly the part we automated. SiteHunt takes a lead list, screens each business for a missing or broken website, builds real sites from templates keyed to the vertical, and helps you send the outreach. Using this pipeline for our own agency, we screened 402 businesses and published 43 sites in a week, which is not a volume any of us would have reached by hand.

SiteHunt is in early access for freelancers and agencies. You bring your own API keys, and the sites you build carry no SiteHunt branding.

The Takeaway

Nobody replies to a stranger asking whether they want to buy something. People do reply when a stranger has already done the work and is holding the result. Build first, then pitch. It costs more up front and it wins far more often.