40 UK home care providers were scored against a published standard. Not one passed a single dimension.

There is a widely repeated explanation for why new home care providers struggle to win private clients: the market is hard, families are loyal to incumbents, and it takes years to build a reputation. Some of that is true. None of it is what we found when we looked at the websites.

By 2026-09-03, 40 UK care providers had received a full technical audit across our campaign and other inbound enquiries. We scored each one against twenty checks grouped into five dimensions — can a family find them, believe them, contact them, get answered by them, and can the provider afterwards say where that family came from.

A dimension passes only when all four of its checks pass. That is a hard bar and it was set deliberately, because the question the standard answers is not is this a good website. It is can this business run paid acquisition without leaking. Those are different questions, and the second is the one that costs money.

Not one provider in the cohort passed a single dimension.

The method is published below alongside the rubric itself, so the result can be reproduced. Where a check could not be observed by the means we used, it is recorded as unknown — never as a failure — and excluded from the denominator. Every figure on this page states the number of providers on which that check could actually be resolved.

What we found

These are the checks that no provider in the cohort passed, ordered by how many providers the check could be resolved on. Each is stated with its own denominator.

Home care enquiries arrive by phone more than by form. Without call tracking or a per-source number, a phone enquiry cannot be traced to the advertising that produced it, and the channel that works looks identical to the channel that does not.

New enquiries arrive in an ordinary email inbox rather than a tracked enquiry pipeline. Without a recorded enquiry state, the team must manually establish which enquiries were answered and which still need follow-up.

The five dimensions

A dimension passes only if all four of its checks pass. A dimension fails if any check fails. If none fails but at least one is unknown, the dimension remains unresolved — neither passed nor failed.

Credibility — 33 of 33 decided providers failed

A family checking a provider is doing basic verification: does the regulator record match, is there a name attached to the business, does anyone independent vouch for it. These are cheap to fix and they are the first thing a self-funding family looks for.

Place — 30 of 30 decided providers failed

Home care is bought locally. A site that does not name the town it serves in the places search engines and families actually read is invisible for the only searches that can produce a client.

Capture — 28 of 28 decided providers failed

The enquiry form is the narrowest point in the whole funnel. Every failure here is paid for twice — once in the advertising that brought the family, and once in the enquiry that never arrives.

Response — 18 of 18 decided providers failed

This is the dimension a website cannot show you, so it was scored from what each provider described on a recorded call. It is the point at which advertising either becomes a client or quietly stops existing.

Measurement — 27 of 27 decided providers failed

Measurement is where the cohort fails most completely. It is also the dimension that decides whether advertising can be improved or only repeated.

The full result table

All twenty checks, all 40 providers, every denominator shown. This is the entire dataset behind every figure on this page. Bars show the share of resolved providers that failed.

Who is in the cohort

40 UK care providers. By nation: 38 England, 2 elsewhere in the UK. By stage: 19 newly registered, 13 established, 4 franchisee, 4 pre-launch. All provide care in people’s homes; several also provide supported living, live-in care or healthcare staffing alongside it. 37 of 40 had a live website at the snapshot date. 30 of 40 were also audited through a recorded discovery call, which is what makes the Response dimension resolvable at all.

The stage mix matters when reading the results. This is not a sample of long-established national brands. It is weighted towards newly registered and pre-launch providers — businesses building their private-client acquisition systems, where enquiry handling and measurement matter from the first campaign.

What one enquiry is worth

This report states no conversion rate, because no defensible published enquiry-to-client conversion rate for UK home care exists. What it can do is price the unit, entirely from published figures, and let the reader supply the rest of the arithmetic.

The Homecare Association — the sector’s own trade body — publishes a Minimum Price for Homecare each year: the rate below which care cannot be delivered sustainably and lawfully. For England in April 2026 to March 2027 that minimum price is £34.42 an hour — a cost floor, not a market rate. Public commissioners across the UK actually paid an average of £24.36 an hour in the April 2025 sample week — a separate benchmark for public funding. These figures put the cost of delivery and the funding environment in view. For providers building a private-client service, pricing and a dependable enquiry journey belong in the same commercial plan.

Every figure below is a published primary source. Nothing here is estimated, modelled or supplied by us.

Input	Value	Source
Minimum Price for Homecare, England 2026/27 April 2026 to March 2027	£34.42 per hour	Homecare Association, response to the Low Pay Commission 2026 Consultation (§2.5.4)
Average price actually paid by public commissioners, UK sample week, April 2025	£24.36 per hour	Homecare Association, The Homecare Deficit 2025
Share of domiciliary care users who fund their own care, England 2022/23	23.5%	ONS, Estimating the size of the self-funding population in the community, England: 2022 to 2023
Active CQC locations registered as homecare agencies, England 26 August 2026	14,501 locations	CQC care directory (derived from the published CSV)

The unit

At the Homecare Association minimum price of £34.42 an hour, a package of 12 to 18 hours a week is worth £413 to £620 a week — on the order of £21,478 to £32,217 a year if it runs a full year.

Every number above is a published figure from a named source. The arithmetic from there belongs to the reader — and it does not take many recovered enquiries a year before the question answers itself.

Why the cohort’s failures are expensive rather than untidy

A provider that fails M2 and M4 has not merely lost some reporting. Advertising platforms optimise delivery towards the event they are sent. Sent no enquiry event, they optimise towards the cheapest proxy available to them — clicks and page views — because that is the only signal they have. The budget still spends. It spends towards the wrong outcome, and increasing it does not correct that.

The same holds at the other end of the line. An enquiry that lands in an inbox with no state (R1) is indistinguishable, a week later, from one that was answered. A provider in that position cannot tell an advertising problem from a follow-up problem, and in our experience concludes that the advertising did not work.

Set against a market of 14,501 active homecare locations in England alone, where ONS put self-funders at 23.5% of domiciliary care users in 2022/23, the competition for the private client is not won by whoever advertises hardest. It is won by whoever can still see what happened.

The standard is published

The rubric used here is The Private Client Readiness Standard, v1.0. It is published in full, with every check and its pass condition written out, and it is free to reproduce and apply with attribution and a link. A standard other people can apply without us is worth more than a finding only we can produce.

Five dimensions, four checks each:

• Credibility — regulator record matches the site, one phone number, a named accountable person, independent review evidence.
• Place — the town is in the title or h1, a real service-area page exists, claimed coverage is credible, a matching local business record.
• Capture — click-to-call works, the form is short, the form learns the four things, submissions are confirmed to arrive.
• Response — enquiries land somewhere accountable, the enquirer is acknowledged automatically, first response has a named owner, a response standard exists.
• Measurement — analytics on every page, a conversion event fires, a confirmation destination exists, calls are attributable.

Each check resolves to pass, fail or unknown. A provider is ready when all five dimensions pass. It is a hard bar on purpose.

What the standard measures

• It measures commercial readiness, not care quality. Nothing here is a view on how well anyone cares for anyone.
• It measures each provider against the published rubric, not against other providers. There is no league table and there will not be one.
• It is an independent commercial assessment and has no standing with any regulator.

Method

How providers entered the cohort

UK home care providers who came in as inbound leads from the CareClients private-client campaign and subsequently received the full 20-check audit. Records were assembled from four independent audit workstreams and deduplicated by provider identity rather than by file count. Providers who received only a light triage rather than the full audit are not in this file and are not counted in any figure on this page.

The campaign produced 72 contact records over the period; 33 of them are in this cohort of 40. It reconciles exactly, and the split is the interesting part. Of 72 campaign contacts, 33 were resolved to a real business and are in the cohort. Only 19 of those arrived on a company email domain; the other 14 came in on a personal address, or on no address at all, and were resolved from the business name in the enquiry, the call notes, or the email local-part. A personal email is therefore not the end of the trail, and we did not treat it as one. A further 12 contacts were resolved to a business and then excluded, each listed with its reason — staffing, recruitment, training and accountancy businesses that do not deliver home care, providers whose identity could not be established at all, and national franchise networks where the enquiry could not be attributed to the one franchisee who sent it. The remaining 27 could not be resolved to a business from anything the contact gave us: no company name, no website, and no register entry that could be matched with confidence. They are not counted as failures, because an unresolved identity is not a finding. The cohort is larger than 33 because 4 providers were audited to the same standard outside this campaign, and for the same reason the exclusion list runs to 13 entries rather than 12. No contact was dropped for scoring badly, and none was dropped for being a lost sale — the disqualified stage of the pipeline is included in full, because a lead lost on price is still a provider with a website, and dropping it would bias the cohort towards the providers who liked our quote. The funnel above describes the CRM export of 30 August 2026 and is left as it stood on that date, because it is a true statement about that export. A call-led backlog run on 3 September audited eight further providers. Three of them were already in the cohort and were revised rather than added. Five were new to it: three were resolved to a regulator-registered home care provider and joined the cohort, and two were resolved to a business and then excluded, each with its reason. One of those five had been discovered before the 30 August export and was among the contacts that export could not resolve; the other four arrived after it. No contact was dropped for scoring badly. 15 further records were audited and then removed: businesses that turned out not to deliver home care, providers whose identity could not be resolved to a single organisation, and national franchise networks where the enquiry could not be attributed to the individual franchisee who sent it. Each is recorded internally with its own reason.

How each check was scored

• Pass — verified true by direct observation of the live site, the public regulator register, or the provider’s own statement on a recorded call.
• Fail — verified false by the same means.
• Unknown — not observable by the method used. Never scored as a fail, and excluded from the denominator.

Every check in the evidence file carries a verbatim observation supporting its score, and every figure on this page is calculated only on the providers where the check resolved. Nothing is inferred, weighted or estimated. Scores are a snapshot at 2026-09-03, and the full table is published so the same twenty checks can be applied to any provider, by anyone, without us.

Cohort and consent

Every provider scored here was audited with its own knowledge, as part of a commercial enquiry it initiated. Published findings are aggregated. Individual providers are never identified, and no combination of published details is permitted to make one identifiable — including page counts paired with a distinctive technical fault. Per-provider scorecards remain confidential to the provider concerned.

---

Citation and reuse

This report is free to quote, reproduce and build on, including commercially, with attribution and a link. No permission request is needed and there is nothing to fill in to read it.

The report is refreshed annually. Figures are versioned by snapshot date so a citation of the 2026 edition stays valid after the 2027 edition publishes.