Web design for an estate agent in Surrey comes down to one page: the one where a homeowner in Weybridge, Woking or Walton-on-Thames asks you to value their house. The portals already own the property search, so your site cannot win on listings. It can win the instruction, if the valuation request is easy to make and is answered within minutes, at any hour. This guide covers what that page needs, what the local search data says, and three honest ways to get there.
The one job your website has left
Rightmove and Zoopla are where buyers look. You pay to be there, every agent in your town pays to be there, and no independent agency site is going to pull buyers away from them. That is not a failure. It just means the "search our properties" section on your homepage is doing far less than it looks like it is doing.
What the portals cannot do is choose the agent. When a homeowner in Esher decides to sell, they search, they shortlist two or three agents, and they ask each for a valuation. The agent whose site makes that request easy, and who rings back first, gets the first appointment. Everything in your fee for the year is downstream of that moment. So the site has one measurable job: valuation requests, and how fast they get answered.
The problem in your money: a worked example
Put your own fee in; this is an example to show the shape of the sum. Say your average fee on a completed sale is £4,000 (swap in your real figure). Say the site brings in four valuation requests a week across your towns, and your history says one in four valuations becomes an instruction, and four in five instructions complete.
4 requests x 0.25 instructed x 0.8 completed x £4,000 x 52 weeks = £166,400 a year
Now the leak. Suppose a quarter of those requests arrive in the evening or at the weekend and get a reply on the next working morning, and that half of those sellers have already booked the first slot with someone who answered sooner:
4 requests x 0.25 out of hours x 0.5 lost = 0.5 requests a week lost 0.5 x 0.25 x 0.8 x £4,000 x 52 = £20,800 a year
That £20,800 is not a design problem. The page may be lovely. It is a response-time problem, and the design and the response have to be built together or the page just generates leads for the agent who picked up. The speed-to-lead guide walks through why the first meaningful reply so often wins; the logic is the same for a valuation as for a boiler.
What the data says: Surrey estate agent searches in 2026
These are UK search volumes from Google Ads data, 2026 run-rate, checked on 13 October 2026. They show two things: estate agency is one of the few local trades with real demand in three systems at once (the site, the response, the follow-up), and that demand is town by town.
| Search | Searches a month (UK, 2026 run-rate) |
|---|---|
| estate agent woking | 708 |
| estate agent surbiton | 678 |
| estate agent kingston | 606 |
| estate agent surrey | 579 |
| estate agent twickenham | 512 |
| estate agent staines | 490 |
| estate agent weybridge | 440 |
| estate agent teddington | 408 |
| estate agent walton on thames | 322 |
| estate agent cobham | 320 |
| estate agent egham | 309 |
| estate agent esher | 226 |
| estate agent addlestone | 171 |
| estate agent chertsey | 165 |
| estate agent shepperton | 165 |
| estate agent sunbury | 132 |
And the terms agents themselves search:
| Search | Searches a month (UK, 2026 run-rate) |
|---|---|
| web design for estate agents | 232 |
| crm for estate agents | 202 |
What that means in plain terms: a seller searches the town, not the county. "Estate agent surrey" at 579 is smaller than Woking alone. An agency covering Weybridge, Walton and Shepperton is looking at more than 900 searches a month across three towns, and a single homepage cannot rank for all three. Each town needs its own page, written for that town, with sold prices from that town and a valuation form on it.
Honest caveat: these are searches for an estate agent, and a share of them are buyers and tenants, not sellers. The valuation page does not need all of them. It needs the sellers, and sellers are the ones who fill in forms.
What the valuation page needs
A valuation page is a single page whose only action is "book a valuation". Here is what it needs, and why.
- The form above the fold on a phone. Postcode, name, mobile, best time. Four fields. Every extra field costs you requests. The mortgage calculator and the team photos go below.
- Proof from that town. Three recent sales in the same postcode area with the price achieved and the days to agree a sale. Real figures only. A seller in Cobham does not care what you sold in Woking.
- Reviews pulled in live from Google, not typed in. A seller is about to trust you with the biggest sale of their life; they will check.
- An instant reply. The moment the form is sent, the seller gets a text confirming it and saying when they will hear from a named person. Then a human, or a receptionist that answers and books on your behalf, rings within minutes, including on a Sunday.
- A CRM behind it, the one list of every seller, buyer and viewing with what happened last, so the request lands on a named negotiator's list with a timer, not in a shared inbox.
Everything else on the site is optional. The listings feed from your portal, the "about us", the area guides: keep them, they do no harm. Just stop judging the site by them.
The three options, honestly
Option 1: keep your current site and fix the valuation form yourself
What it means: no rebuild. You cut the form to four fields, add three real sold prices to the page, set up an automatic text reply, and agree a rota so someone rings back within fifteen minutes during opening hours. How you know it is working: count requests and time-to-first-call for a month. Where it falls down: evenings and weekends still go to the inbox, and if your site is on a platform your portal supplier controls you may not be able to edit the page the way you need. Who it is right for: a one-office agency that already gets requests and mostly loses them on response time. Do this first regardless. It costs an afternoon.
Option 2: an estate-agency template platform
What it means: one of the industry suppliers that provides a site with the portal feed built in, a standard valuation form and a standard design, for a monthly subscription. How you know what you are buying: ask for three live sites on the platform in Surrey and search their towns; see whether the sites rank or the portals do. Where it falls down: the pages are the same across hundreds of agencies, so Google has no reason to rank yours for Weybridge above the next one; the valuation form goes to email; and there is usually no out-of-hours answer. Who it is right for: an agency that mainly needs a tidy listings site and gets its instructions from reputation and the high street. If that is you, this is cheaper than a build and good enough.
Option 3: a valuation-led site with instant response and CRM follow-up
What it means: one real page per town you sell in, each with that town's sold prices, live reviews and the four-field form; an instant text reply and a call back within minutes at any hour, by your team in hours and by a receptionist that answers and books outside them; every request logged in a CRM that nudges the negotiator and re-contacts the seller who said "not yet". How you know it works: you see requests per town, minutes to first call, and instructions won, every month. Where it falls down: it costs more than a template and takes weeks; it is overkill for a single-office agency that is already winning every valuation it is asked to do. Who it is right for: an agency covering several towns, losing valuations to faster rivals, with more than one person handling leads.
The recommendation
Bottom line: do option 1 this week whatever else you decide; then, if you cover more than one town and your out-of-hours requests go unanswered until Monday, option 3 pays for itself on the first instruction it saves.
The reason is in the sum. In the example above, the response-time leak alone was worth £20,800 a year on a £4,000 average fee. One extra completed instruction covers most builds. A template platform does not close that leak because it does not change who answers at 8pm on a Saturday. If you are a single office and you answer every request inside the hour, you do not need option 3, and anyone selling it to you is selling you a website rather than a result. What we build is the site; the receptionist is the answer at 8pm; the CRM is the follow-up. Buy the parts you are missing, not the set.
How to do it this week
- Time your last ten valuation requests. Open the inbox or CRM. For each request note the time it arrived and the time someone first rang. Write the gap down. Count how many arrived outside 9am to 6pm or at the weekend. That is your leak, in requests.
- Price the leak. Average fee x your instruction rate x your completion rate x requests lost. Use your own figures. If the number is under what an afternoon of your time is worth, stop here and fix the rota.
- Cut the form. Open your valuation page in the site editor. Remove every field except postcode, name, mobile and best time to call. Done looks like: four fields, the "book my valuation" button visible on a phone without scrolling.
- Add three local proofs. On each town page, add three sales from that town in the last six months: road name, price achieved, days from listing to sale agreed. Real figures from your own records.
- Set the instant reply. In your form tool or CRM, switch on an automatic text to the seller's mobile the moment the form is sent: "Thanks [name], [negotiator] will call you within the hour / first thing tomorrow about [postcode]." Done looks like: you submit a test request from your own phone and the text arrives within a minute.
- Set the rota and the timer. Name one person per day who owns valuation requests. In the CRM, set a task that fires fifteen minutes after a request if nobody has logged a call. If you have no CRM, use a shared calendar alert; it is clumsy but it works for a week.
- Check your town pages exist. In a private window, search "estate agent" plus each town you sell in. If your site is not on page one and a page for that town does not exist, that page is the next thing to build. Add each town page to your Google Business Profile service areas as well.
- Measure for 30 days. Requests per town, minutes to first call, valuations booked, instructions won. Then decide about option 3 with a real number in front of you.
Mistakes we see
- Judging the site on the listings. The portal feed looks like the site's purpose. It is not. Sellers, not buyers, pay your fee.
- A twelve-field valuation form. Bedrooms, property type, reason for selling, timescale. Ask those on the phone. Every field on the page costs you requests from people on a mobile.
- The form goes to info@. Shared inboxes are where valuation requests go to die over a weekend. One named person, one timer.
- One homepage for five towns. It ranks for the brand name and nothing else. Weybridge sellers search Weybridge.
- Dropping the "not yet" sellers. A seller who asked for a valuation and did not instruct is the warmest lead you own. Six months later they are still the warmest lead you own, if anyone remembers to call. That is what the CRM follow-up is for.
The decision
If you cover more than one Surrey town and you want to know how many of your valuation requests are leaking on response time before you spend anything on a rebuild, send me your last month's requests with timestamps and I will come back with the gap, in requests and in fee. Want me to run that? If every request already gets a call inside the hour, say so and I will tell you to keep your current site and fix the form yourself.




