Everything in this guide is public information. We have no relationship with any business named here. We compete with most of them. This is our attempt to gather every public data point we could find and turn it into something useful.
We are not saying the top 10 are all good, and we are not saying an agency outside the top 10 is bad. We are not endorsing any agency in this list, nor are we discounting any that have not made it. A score is a summary of public signals on the dates stated, not a judgement of the people, the quality of their work or how they would perform for you. You should enquire with whichever agencies appeal to you most, get multiple quotes and make a call based on your own assessment.
We are an SEO agency, and we are in this list. Local Digital built the index, paid for the data and is scored by the same rules as everyone else. Our row is marked, our full scorecard is published, and the weights were fixed before we ran the numbers. Every figure comes from public sources: Google Business Profiles, Trustpilot, Glassdoor, LinkedIn, ABN Lookup, the Wayback Machine, each agency's own website and DataForSEO's keyword database. No agency paid to be listed or to move.
The limits. Public data can be incomplete, out of date or wrong, and third-party estimates such as keyword databases are estimates. Figures are correct to the best of our knowledge on the read dates shown and will change when we re-run the index. Nothing here is professional, financial or legal advice, and we accept no liability for decisions made on it. If your agency is named and a figure is wrong, contact us: we correct and log every change, and every agency gets a right of reply before the index is re-run.
What this page is
This is the complete rulebook for the Local Digital Agency Index, the scoring behind Best SEO agencies in Australia and its eight city pages. Every pillar, sub-metric, threshold and source is here, so every score on those pages can be traced to a rule. Weights are frozen for 12 months from September 2026. The index is re-run periodically, with a public change log on this page.
Step 1 · Building the candidate pool
232 agencies across nine markets (Sydney, Melbourne, Brisbane, Perth, Adelaide, Gold Coast, Canberra, Newcastle and the Sunshine Coast) came from seven sources, de-duplicated by domain, business name and Google place ID:
- The organic top 30 for “seo agency [city]” and “best seo agency [city]” in every market, and the local packs for the same terms.
- A depth-100 Google Maps sweep across 32 city and suburb queries (1,619 places, filtered to 15 or more reviews and a marketing or SEO category).
- Clutch and Semrush Agency Partners listings for Australia.
- The competitor listicles and directory pages that rank for each city term.
- The answers of three AI engines (Perplexity, ChatGPT, Gemini) to “best SEO agency in [city]”.
- Named nominations from our own team, which is how several agencies we compete with day to day were added.
Step 2 · Eligibility and listing rules
- Not an SEO agency by evidence. Excluded when under 20% of reviews are about SEO and the Google category is web, design or branding; or when under 15% of 20 or more text reviews are about SEO whatever the category. One override: an agency that ranks in the organic top 10 for a city “seo agency” term is listed and its low SEO share simply scores low.
- Educator or consultant. Excluded when a majority of reviews are for workshops, webinars, courses or advisory. A quarter earns a card note, not exclusion.
- Listing thresholds. A verified Google Business Profile with 5 or more reviews in the city. 5 to 19 reviews is listed with a thin-review-base note. No profile in the city means unlisted.
- Integrity tests. An agency is analysed and named in the candidate list but not ranked when two or more of five evidence tests are met with dated citations: three or more independent-platform complaints of billing after cancellation or withheld assets in 24 months; owner replies that deny the reviewer or threaten legal action on two or more negative reviews; a review-suppression allegation against an implausibly clean profile; a “no lock-in” claim contradicted by the published terms; an independent negative share of 15% or more over 20 or more reviews. The tests are published here; no agency is named against them, and each has a right of reply.
- Boutique. Fewer than 10 staff on LinkedIn. Scored and shown on the city list, not ranked in the national index, because a founder-led team of six and a 50-person agency are different purchases.
- National index. Every listed agency with 20 or more Google reviews that is not a boutique. Agencies with 5 to 19 reviews are ranked on their city page only.
Step 3 · The nine pillars and their sub-metrics
| Sub-metric | Points | Rule | Source |
|---|---|---|---|
| A1 Google rating | 8 | 5.0 → 8 · 4.9 → 7 · 4.8 → 6 · 4.7 → 4 · 4.6 → 3 · below → 1 | Google Business Profile |
| A2 Review volume | 2 | ln(count) ÷ ln(400) × 2, capped at 2 | Google Business Profile |
| A3 Negative rate | 6 | Lifetime 1–2★ share: ≤1% → 3 · ≤2% → 2.5 · ≤3% → 2 · ≤5% → 1. Last 24 months: 0% → 3 · ≤3% → 2.5 · ≤6% → 2 · ≤10% → 1 | Full Google review history |
| A4 Independent platforms | 6 | 10 or more Trustpilot reviews: negative share ≤3% → 6 · ≤8% → 4 · ≤15% → 2; minus 2 if the Google rating exceeds the independent rating by 0.5. Under 10 reviews → neutral 4 | Trustpilot |
| A5 Review integrity | 6 | Start at 6. Review drive on the full history (−2 when 15% or more of lifetime reviews land in 30 days, or 15 or more in a week; −1 at 10% or 10) · −1/−2 when non-client reviews exceed 12%/25% · −1 when blank reviews exceed 20% · −1/−2 for owner replies that dispute the reviewer | Review text, dates, replies |
| B1 SEO share of reviews | 10 | Share of reviews that are about SEO × 10 | Review text |
| B2 Positioning and capability | 4 | Positioning × 0.6 (SEO-only 3 · SEO-led 2.4 · generalist 1.2 · web 0) + verifiable partner status × 0.5 | Agency site, partner directories |
| C1 Organic rank | 7 × 8⁄13 | Best of home-city and national terms on the snapshot day: top 3 → 7 · 4–6 → 5 · 7–10 → 3 · 11–30 → 1 | Google SERP, same day for all |
| C2 Local pack | 4 × 8⁄13 | Top 3 in the local pack for the home-city term → 4 · Maps top 20 → 2 | Google SERP, Maps |
| C3 AI visibility | 2 × 8⁄13 | Named by 3 engines → 2 · 2 → 1.5 · 1 → 1 | Perplexity, ChatGPT, Gemini |
| D Tenure | 10 × 0.8 | Years since founding: 15+ → 10 · 10–14 → 8 · 7–9 → 6 · 4–6 → 4 · under 4 → 2. Earliest of ABN active-from date, first Wayback capture, first Google review, or a corroborated stated year | ABN Lookup, Wayback Machine, Google |
| E Own-domain authority | 8 × 0.75 | (ln(referring domains) ÷ ln(2500) × 7 + spam band) × 0.8 | DataForSEO backlinks |
| F1 Reply rate | 2 | Share of reviews with an owner reply: 80%+ → 2 · 50%+ → 1 | |
| F2 Reply quality | 2 | Replies to negative reviews that are accountable rather than dismissive | |
| F3 Published pricing | 2 | Any published price for the agency’s own SEO, whether a package, a from-price or a range → 2 · nothing or quote-only → 0 | Agency site |
| F4 Published contract terms | 3 | Published month-to-month → 3 · published minimum of 3 months or less → 2 · unpublished, 6–12 month minimum or 60-day notice → 1 · balance-of-term liability or a “no lock-in” claim contradicted by the terms → 0 | Agency terms page |
| F5 Honest claims | 1 | No ranking or traffic guarantee on the site | Agency site |
| G1 Owned-media output | 3 × 6⁄7 | Blog posts in 12 months (24+ → 1.5 · 12+ → 1 · 4+ → 0.5 · dormant 0 · no dated source → neutral 0.5) + podcast or video episodes in 12 months (12+ → 1.5 · 4+ → 1 · 1+ → 0.5) | WordPress API, RSS, Apple Podcasts, YouTube |
| G2 Research, tools, events | 2 × 6⁄7 | Original study 1 · curated statistics 0.5 · free tool 0.5 · conference or training 0.5, capped at 2 | Sitemap review |
| G3 External recognition | 2 × 6⁄7 | Independent awards (2+ → 1 · 1 → 0.5) + headline-verified press in 12 months (3+ → 1 · 1+ → 0.5) | Award bodies, Google News |
| H1 Employer rating | 5 | URL-verified Glassdoor, Indeed or Seek profile: 4.5+ → 5 · 4.0+ → 4 · 3.5+ → 2.5 · 3.0+ → 1; small samples pulled toward neutral 3; no profile → 3 | Glassdoor, Indeed, Seek |
| H2 Tenure and scale | 5 | Under 10 staff → 0 (boutique) · 10–19 staff capped at 3 · 20+ staff: median tenure 3+ years → 5 · 2.5+ → 4 · 2+ → 3 · 1.5+ → 2 · 1+ → 1; unread → 3 | LinkedIn Premium insights, read by hand |
| H3 Turnover complaints | −2 | Client reviews citing account-manager turnover: 2–3 → −1 · 4+ → −2 | Google and Trustpilot text |
| I0 Client selection | — | Up to three named clients from the agency’s own case-study, results or portfolio pages: SEO case studies first, newest first; ads-only case studies, global brands without an Australian site and government bodies skipped; name resolved to a domain by Google search and checked by hand. No named client anywhere on the site → 0. Not audited (outside every top 10) → the audited median, so the pillar is neutral | Agency site, Google |
| I1 Visibility level | 1 per client | Keywords in Google Australia’s top 10: 2,000+ → 1 · 500+ → 0.75 · 100+ → 0.5 · 20+ → 0.25 | DataForSEO Labs |
| I2 Trend against the market | 2 per client | Change in estimated traffic value and keyword count, July–September 2026 against July–September 2025, minus the median change across all audited clients (−24%): 30+ points above → 2 · 10+ → 1.5 · within 10 → 1 · within 30 below → 0.5 · further below → 0. Fewer than 20 keywords or under 100 in traffic value in the base period, or under 10 top-10 keywords now → neutral 1 | DataForSEO Labs, 16-month history |
| I3 Strength | 1 per client | Share of ranked keywords in the top 3: 25%+ → 1 · 15%+ → 0.5 | DataForSEO Labs |
Pillar totals: A verified client sentiment 28 · B SEO focus 14 · C own rankings 8 · D tenure 8 · E own-domain authority 6 · F transparency 10 · G thought leadership 6 · H team stability 10 · I client results 10. The C, D, E and G sub-metrics are scored on their original scale and multiplied by the factor shown so the nine pillars add to 100.
Step 4 · What was read, and how
- Google reviews. The full review history of every listed agency, not the first page: 209 profiles and more than 15,000 reviews through the DataForSEO Business Data API. Ratings and counts are as at 14 September 2026.
- Trustpilot. Looked up by domain for all 232 agencies. Australian SEO agencies have no ProductReview presence; the six raw matches there were other businesses with the same word in their name and were discarded.
- Employer ratings. Glassdoor, Indeed and Seek, found through Google’s rich results and then verified by the profile URL, because a name search alone returns the wrong company for generic names.
- Headcount and tenure. LinkedIn Premium company insights, read by hand and recorded with the date. 21 agencies have been read; the rest are neutral on H2 until the next read, and the tables say “not read”.
- Rankings. One Google SERP snapshot for every agency, currently 21 September 2026 (first read 14 September), because positions move materially between days. AI-engine mentions from stored answers on the same week.
- Tenure. ABN Lookup active-from date, first Wayback Machine capture, first Google review, and stated founding years where two sources agree.
- Terms and pricing. Each site’s pricing and terms pages, found by search and sitemap; the terms score uses the published document, not the sales page.
- Publishing output. WordPress REST API or RSS for dated posts, Apple Podcasts and YouTube for episodes, sitemaps for research and tools, award bodies and Google News for recognition. Sitemap dates were unreliable and were not used.
- Client audit. DataForSEO Labs, Australia, one pull on 15 September 2026: keywords ranked, top-3 and top-10 counts, estimated traffic value and monthly history from June 2025 for 188 client domains. These are modelled estimates from a keyword database, not Google Search Console data, and the same yardstick for everyone. Across 159 domains with a meaningful base period the median change was −24% and 77% fell, which is why trend is scored against the market.
Step 5 · Ranking, disclosure and corrections
- The national index ranks every listed agency with 20 or more Google reviews, boutiques aside, whatever its city. City pages rank every eligible agency with a verified office there and show boutiques in their own group.
- Every table shows the nine pillar scores, the Google rating and count, headcount and tenure where read, and the audited clients with their numbers.
- Local Digital is the author, the funder and a scored participant. Our row is marked on every page, our weights were fixed before the numbers were run, and we sit where the rules put us.
- No agency paid to be listed or to move. Every named agency has a right of reply before each re-run, and every correction is logged below with the date.
Known limits
- Agencies choose their own showcase clients, so the client audit measures the best work they will show, not the average client. Big brands earn the visibility point regardless of the agency, which is why that part is capped at 1 of 4.
- Employer ratings and LinkedIn tenure are neutral where no verified profile or read exists, which favours neither side but does compress the field.
- Thought leadership counts what is measurable: dated posts, episodes, research, tools, awards and press. Social posting and follower counts are deliberately not counted.
- A migration or rebrand mid-window breaks a client’s trend either way; where we found one, the client was dropped and the reason recorded.
- The keyword database and Google itself changed materially during the year. That is exactly why the numbers are dated and re-run periodically.
Change log
- 21 September 2026. Own-rankings pillar refreshed for every ranked agency on a 21 September SERP snapshot; 29 agencies moved. Local Digital re-read on its live site.
- 16 September 2026. First publication. 232 agencies pooled, 134 listed, nine pillars, 188 client domains audited.
Questions people ask
Why are the weights what they are?
Sentiment is the largest pillar because a full review history is the most direct evidence of what clients experience. Client results is the newest and is weighted above the agency's own rankings because a buyer is hiring results for their site, not the agency's. Weights are frozen for 12 months so scores can be compared over time.
Why does an agency with no case studies score 0 on client results?
Because the pillar measures evidence the agency itself chooses to publish. An agency that shows no client is asking buyers to take results on faith, and the score says so. It can change that at the next re-run by publishing named work.
How do I correct a number?
Contact us with the agency name, the figure and the source. We correct, log the change with the date on this page, and every named agency gets a right of reply before each re-run.
