Eight in ten Irish accountancy practices cannot be found in their own town

1,956 of 2,341 practices are missing from their own town's page one

For each of the 122 towns with three or more censused practices, the query accountant <town> was run from the town's own centroid, depersonalised. A practice counts as ranking if its registered domain appears in the top ten organic results.

385 practices rank. 1,956 do not. A further 67.1% do not appear in the local map pack either — the three-firm block that sits above the organic results.

Practices with no website are counted in the denominator and recorded as not ranking, because that is the substantive reality for someone searching.⁴

The three measures where your county sits furthest from the national figure

Aggregate only — no practice can be looked up, by name or otherwise. Percentage-point gaps are against the national figure for the same measure.

Half the profession has never been reviewed once

1,411 of 2,775 practices have no Google reviews at all. Another 956 have between one and four. The median practice has zero; the mean of 5.7 is carried by a small number of firms with a lot of them.

15,700 reviews are shared between the whole profession. 45 practices — 1.6% — hold fifty or more each.

Review counts are published in bands in the open dataset, so no practice can be identified from them.

All 2,775 censused practices, July 2026. Median 0, mean 5.7.

Each square is one per cent of 1,582 live sites.

All measured on the homepage only, so every figure is an upper bound on absence — work that lives on deeper pages is not counted.⁷

231 sites block AI crawlers. Eight of them decided to.

14.6% of the 1,582 live sites disallow at least one named AI crawler in robots.txt. Read as professional intent, that would be a story about a cautious profession. It is not one.

223 of the 231 block the identical set of seven agents, and every one of those 223 is served by Cloudflare — the signature of a managed AI-blocking default, not 223 independent decisions. Strip the template and eight firms remain: 0.5% of live Irish accountancy sites.

A further 18.9% publish no robots.txt at all, which grants access by omission rather than by choice.⁸

231 sites blocking a named AI crawler, of 1,582 live sites crawled 29 July 2026.

Dublin is not ahead. Invisibility is national.

We expected a regional lag and did not find one. 13.5% of Dublin's 945 practices are completely invisible, against 16.5% outside Dublin — a gap of three percentage points, not a chasm. Galway, with 185 practices, outperforms Dublin on every measure in this chart.

The spread that matters is not urban against rural. It is the tail: Roscommon, Tipperary, Laois and Louth sit above 21%, and Waterford sits at 8.6%.

We report this because it is what the data says, and because a regional-lag story would have suited us better.

Share of censused practices with no claimed profile, no website and no reviews. Scale 0–25%.

How this was measured, in 150 words

Practices were sourced from Google's business-listing data via DataForSEO, queried as 108 overlapping 30km circles covering the Republic of Ireland across ten accountancy categories, then deduplicated on place ID: 6,829 raw records became 2,775 practices. Counties were resolved in three passes — Google's own region field, a hand-checked town lookup, then k-nearest-neighbour on coordinates for the remaining 3% (98.0% accurate on a 300-firm holdout).

Every practice listing a website received exactly two requests: its homepage and its robots.txt, from an identifying user agent. No forms were submitted, no authenticated area was accessed and no access control was circumvented. For each of the 122 towns with three or more censused practices, accountant <town> was run from the town's own centroid, depersonalised, desktop, to depth 20.

The limits of this study, stated where they belong rather than at the bottom of a PDF.

1. 12,775 practices is not a claim to have surveyed every practice in the Republic; no complete public list exists. It is the deduplicated set discoverable in Google's business-listing data on 29 July 2026 — which is the population that matters for a study about online visibility, but which cannot include a practice with no listing at all.
2. 2There is no “percentage with no Google Business Profile” figure in this report, because that quantity is not measurable from a Google-derived sample. The measurable figure is “never claimed” — a listing exists, often auto-generated, and no one has verified it: 42.9%.
3. 3No AI Overview appeared on any of the 122 town searches. Because a null result can also mean “not detected”, detection was validated on two informational control queries run through the identical method; both returned an AI Overview block. The absence on local searches is an observation, not an artefact.
4. 4Practices with no website are kept in the ranking denominator and recorded as not ranking, because that is the substantive reality for the person searching.
5. 5The median town is mediocre rather than catastrophic — three of its ten organic results are local practices. The disaster is in the tail: nine towns return none. Showing only the tail would overstate the finding.
6. 6County figures are reported only where at least twenty practices were censused. Leitrim (13 practices) appears on the map for completeness and is excluded from search-visibility comparisons.
7. 7Website measurement is homepage-only, so every “missing” figure is an upper bound on absence. A firm with a pricing page three clicks deep counts here as having none.
8. 8231 sites block a named AI crawler, but 223 of them block an identical set of seven agents and every one is served by Cloudflare — a managed default, not a decision. Only 8 firms show a deliberate AI policy. The raw 14.6% must not be reported as professional intent.
9. 9The accessibility figure was corrected before publication from 86.9% to 52.9%: the first crawl missed the most common correct labelling pattern. All 1,582 sites were re-crawled. A conservative floor of 27.9% is published alongside it, and neither is used alone.

Corrections policy: any error identified will be published on the methodology page. Requests to hello@webnua.com.

What the data says about a single practice

The uncomfortable reading of this report is that most Irish practices are invisible. The useful reading is the other side of the same number: in the median town, seven of the ten page-one results are not a local practice, and the three map-pack places are contested by almost nobody.

A profession where half the firms have never been reviewed and 95.6% publish nothing a machine can read is not a competitive field. It is an empty one.

This report does not rank, name or score any individual practice, and none of the above is advice about a particular firm. It is a description of what was measured across 2,775 of them.

Where does your practice sit in this data?

The report is aggregate by design, so it cannot tell you. A Visibility Audit runs the same twenty measures against your own practice and your own town: live map positions, what page one looks like for your town, and the three things worth fixing first.

If you already hold one of the three map-pack places in your town, you do not need us — and we will say so.