Back to Blog

How to Find Businesses With Outdated Websites

Ilyas Yıldırım
Ilyas Yıldırım
8 min read

Businesses with no website are the obvious prospect for anyone selling web work. They are also the smallest and most heavily worked segment of the market.

The larger opportunity is the business that already has a site, paid for it once, and has not touched it since. That prospect has proven they will spend money on a website, which is the single hardest thing to establish about a cold lead.

This guide covers how to build that list without opening a few hundred sites by hand.

The pool is much bigger than the no-website pool

We classified the website field on a sample of 3,147 local-business listings across 20 countries and 58 cities. The destinations broke down as:

Website field points atRecordsShare
Nothing, field is empty84927.0%
A social or messaging domain1845.8%
A hosted profile, booking page, or short link70.2%
An owned-domain-like destination2,10767.0%

That last row is your addressable market, and it is about two and a half times the size of the no-website segment. It is also the row nobody is competing for, because "they have a website" reads like a disqualification until you look at the sites.

Worth stating plainly: an owned-domain-like destination is a residual class, not proof of ownership or quality. It means the URL did not match our social, booking, or shortener lists. Some of those sites are excellent. Finding the ones that are not is the job.

What you can detect in bulk, and what you cannot

This distinction is where most redesign-prospecting advice quietly overpromises.

Collectable at scale. What platform the site runs on, and what other technologies it loads. This is reliable because it reads what the page actually serves. In a BasedOnBusiness export it arrives as two columns, cms and technologies, filled in by enrichment reading the business website.

Not collectable at scale. Whether the design looks like 2014, whether the copy is any good, whether the booking flow works on a phone, whether the photos are stock. These are the things that actually sell a redesign, and no export contains them.

Be clear about what the enrichment does not do, so you do not build a workflow on a promise it never made: it does not score page speed, test mobile responsiveness, or check certificate state. When you need those, run them separately with a page-experience tool on the shortlist you have already narrowed.

So the honest shape of this workflow is a funnel. Bulk data takes a category from a few hundred listings to a few dozen candidates. A human takes those few dozen to a call list. Anything that claims to skip the second step is selling you an automated PDF that your prospect will recognise as automated.

Building the list

  1. Export a category and area. Same as any prospecting pull: pick the category, pick the location, run it. If your service area is not one city, see scoping a list to any area.
  2. Keep the rows that have a website. This is the inverse of the no-website filter. Everything with an empty website field belongs on a different list with a different pitch.
  3. Read the cms and technologies columns. Group the export by platform. You are looking for clusters, not individual verdicts.
  4. Sort by review count, descending. A busy business with a tired site is the prospect. A quiet business with a tired site is a hobby.
  5. Open the top 20 by hand. On a phone, not a desktop. This is the step that produces the thing you will actually say on the call.

Step 5 is not optional and it is not slow. Twenty sites at thirty seconds each is ten minutes, and it is the difference between "I noticed your site could use a refresh" and "I noticed your booking form does not submit on iOS".

Reading the platform column

The platform a site runs on is a hypothesis about the business, not a verdict on the site.

A site on a mainstream CMS tells you there is probably a theme and plugins behind it, that someone technical was involved once, and that the maintenance question is live. A site on a hosted website builder tells you the owner or a family member most likely built it themselves, that the budget was near zero, and that the conversation is about whether the business has outgrown it. A site running nothing recognisable at all is often the most interesting case, because hand-built static pages in a local category usually mean a developer who has since disappeared.

None of those are pitches on their own. They are how you decide which twenty to open.

Two mistakes to avoid. Do not treat any platform as inherently bad, because the person you are calling may have chosen it deliberately and you will lose the room. And do not read the technology list as a security assessment. Detecting what a page loads is not the same as detecting what is out of date, and telling a stranger their site is insecure based on a detection string is a bad opening and frequently wrong.

Qualifying without doing free work

The temptation with this segment is to send an unsolicited audit. It converts poorly and it costs you an hour per prospect.

What works better is one verifiable observation. You looked at the site on a phone, one specific thing was broken or missing, and you are asking whether that is deliberate. It is short, it is true, it is obviously not a template, and it does not hand over your deliverable before anyone has agreed to talk.

Rank the shortlist on how much the gap is costing them rather than how bad the site is. A plumber with 300 reviews and no way to request a callback is losing money every week. A gift shop with a dated but functional site is not, and will tell you so.

The phone-first structure for these calls is the same one covered in selling web design to local businesses, and the same qualification discipline applies to any local list, which we broke down in the lead list quality scorecard.

Contacting them

Unlike the no-website segment, this one is emailable. These businesses have a website, which is what email enrichment reads, so exports can carry a contact address alongside the platform columns. The mechanics and the limits are in getting emails from Google Maps.

Keep in mind that collecting business contact details and using them for outreach are governed separately, and marketing email in particular has its own consent rules depending on where the recipient is. Our legal and privacy guide covers the shape of that.

FAQ

How do you find businesses with outdated websites at scale? Export every business in a category and area, keep the ones that do have a website, then enrich those records to read what platform each site runs on. That narrows a few hundred listings to a shortlist you can open by hand. The platform and technology signals are collectable in bulk; visual age and mobile behaviour are not, so a human pass stays in the loop.

Is the outdated-website pool bigger than the no-website pool? Considerably. In our 3,147-listing sample, 27.0% had no website link at all while 67.0% pointed at an owned-domain-like destination. The redesign pool is roughly two and a half times the size of the no-website pool, and every business in it has already paid for a website once.

Can a tool tell me if a website is outdated automatically? Only partly, and you should be sceptical of anything claiming otherwise. Platform and technology detection is reliable because it reads what the page actually loads. Judgements like dated design, weak copy, or a bad conversion path are not machine-readable. Treat automated signals as a filter, never as the audit.

What does the technology enrichment actually return? It reads the business website and reports a CMS or platform value plus any other technologies it recognises. Those land in the export as the cms and technologies columns. It does not measure page speed, mobile responsiveness, or certificate state, so pair it with a page-experience tool when you need those.

Which local categories are worth targeting for redesigns? Categories where a site exists but rarely earns attention. Trades, auto, independent food service, and single-location clinics tend to have sites built once and never revisited. Review count is the better filter though: a busy business with an old site has both the need and the budget.

Should I send a free audit as the opener? A short, specific observation outperforms a generic report. Naming one true thing you can verify, such as the platform the site runs on or a form that fails on a phone, is more credible than an automated PDF, and it does not commit you to unpaid work before there is a conversation.

Get started

At basedonb.com, pull a category and area, keep the listings that have a website, and let enrichment fill in the cms and technologies columns before you open a single tab. A new account includes 50 one-time export credits with no card. Paid plans are billed on day one and carry a 7-day money-back guarantee on the value of unused credits.