Methodology

How the Hub Score works

100 points across five factors, every one computed from a fact you can verify yourself. No star ratings we invented, no anonymous testimonials, no paid placement.

Companies scored

79

Countries live

4

Average Hub Score

80/100

Paid placements

0

The five factors, and the exact arithmetic

Each factor has a hard cap, so no single attribute can carry a listing. The caps sum to 102, and the total is then capped at 100 — so a company must be strong on at least four factors to reach the high 80s. The arithmetic below is the arithmetic in the code; nothing is applied by hand.

Tenure

max 25 pts · 25% of score
Formula
round(years trading ÷ 25 × 25), capped at 25. 25+ years = full marks.
Why it is weighted this way
Surviving multiple algorithm eras and at least one recession is the most reliable public signal that delivery is repeatable rather than lucky.
Data source
Founding year from the company's own site, companies register or archived filings.

Service breadth

max 20 pts · 20% of score
Formula
3 points per in-house discipline, capped at 20 (7+ services = full marks).
Why it is weighted this way
Each discipline you have to buy elsewhere adds a hand-off. Breadth is not automatically better, but it is a fact buyers price in.
Data source
Service pages on the company's site; only disciplines delivered in-house are counted.

Vertical depth

max 20 pts · 20% of score
Formula
4 points per industry served, plus an 8-point focus bonus when the company serves 2 or fewer. Capped at 20.
Why it is weighted this way
A clear position scores well at both ends: deep specialists know a vertical's SERP, broad shops carry pattern library across many. Vague middle ground scores lowest.
Data source
Published client work, case studies and stated vertical focus.

Pricing transparency

max 20 pts · 20% of score
Formula
8 points for a monthly-retainer model (6 for other models) + 7 for a stated retainer floor + 5 for a live public site.
Why it is weighted this way
Agencies that will not name a number until the third call cost buyers weeks of qualification time.
Data source
Published pricing pages, disclosed retainer floors, and the live site.

Delivery scale

max 17 pts · 17% of score
Formula
Team size mapped to a curve: 10-20 → 8, 20-50 → 11, 50-150 → 14, 150-250 → 16, 250-500 → 17, 500+ → 15.
Why it is weighted this way
The curve peaks in the mid-hundreds: enough bench depth to survive staff churn, small enough that senior people still touch the account. Very large shops lose a point for delivery distance.
Data source
Headcount bands from the company's site and public professional profiles.

A worked example: WebFX

Same inputs, same output, every time. Here is how one live listing reaches its score.

FactorInputPoints
TenureFounded 1996 → 30 years trading25/25
Service breadth7 in-house disciplines × 320/20
Vertical depth4 industries × 416/20
Pricing transparencyMonthly retainer + floor of $3k/mo + live site20/20
Delivery scaleTeam size 500+15/17
Hub ScoreSum, capped at 10096
See the WebFX profile →

Reading a score in context

Bands are guidance for shortlisting, not grades. A 68 with the right vertical experience beats an 88 that has never worked in your market.

  • Top decile85+

    Long tenure, broad in-house delivery and open pricing.

  • Strong75+

    Established, with either deep focus or wide capability.

  • Solid65+

    Credible operator; expect a gap in tenure, breadth or disclosure.

  • Emergingunder 65

    Younger, smaller or less transparent on price — not automatically worse, but verify more.

What the score deliberately ignores

Stated plainly so nobody assumes these are secretly weighted.

  • Payment of any kind. There is no paid placement, no sponsored row and no pay-to-be-reviewed tier.
  • Star ratings, anonymous testimonials and review-site averages, which are trivially gamed.
  • Traffic or ranking estimates for the agency's own site — an agency ranking for its own brand tells you nothing about your market.
  • Awards and badges, which are largely self-nominated and often paid.
  • Client logos, unless the engagement is documented in a case study with dates and scope.

Limits worth stating

The Hub Score is not a performance guarantee and not a review. It measures how a company presents itself against evidence anyone can check: how long it has operated, what it does in-house, who it serves, how openly it prices and how deep its bench is. It cannot tell you whether a specific agency will rank your specific site.

Every listing is tied to the country and state the company actually operates from, so a region page only shows companies genuinely based there — not distant agencies claiming every city.

SEO Companies Hub is independent. Listings are editorially compiled from public information and cannot be bought. Companies do not pay for placement or for a higher Hub Score.

Methodology questions

Can a company pay for a higher Hub Score or a better position?
No. The score is computed from stored, verifiable attributes and the ordering is the score. There is no sponsored placement, no paid review tier, and no commission on introductions — so there is no mechanism through which money could move a listing.
Why is a 20-person agency sometimes ranked above a 400-person one?
Delivery scale is worth at most 17 of 100 points. A boutique with 20 years of trading, five in-house disciplines, a published retainer floor and a tight vertical focus will out-score a large generalist that hides its pricing. That is intentional: size is one input, not the verdict.
Does a high Hub Score mean the agency will rank my site?
No. The score measures verifiable operating facts — tenure, in-house breadth, vertical focus, pricing openness and delivery scale. It cannot predict performance on your domain, in your vertical, with your budget. Use it to build a shortlist, then interrogate the shortlist.
How do you handle agencies that refuse to publish pricing?
They lose the seven points tied to a stated retainer floor, and two more if their commercial model is not a straightforward monthly retainer. They are not excluded — plenty of good firms price bespoke — but buyers pay for that opacity in qualification time, so the score reflects it.
What triggers a re-score?
Any change to a stored attribute: an acquisition, a headcount band shift, a new in-house discipline, a published price change, or a correction we accept. Tenure re-scores automatically each calendar year, so scores drift upward slightly as companies age.
How do I get a listing corrected or added?
Send the correction with a public source we can check — your own site, a register entry, or an archived page. We do not accept score changes on request without evidence, and we do not remove accurate listings.