Methodology

How every ranking on this site is decided.

For a typical link-building buyer considering this provider today, how strong is the overall purchase proposition across deliverable quality, value, protection, verified customer experience, managed help, commercial flexibility, reporting, and continuity?

The score compares the purchase proposition, not guaranteed SEO results.

Every ranking publishes its criteria, weights and evidence dates.

Methodology versionM1.3 Evidence cutoff2026-08-30 Statuslocked for the 2026-08-30 publication wave

What every ranking must show:

Provider Buyer Score (PBS) one number, 0–100, to one decimal place. Every provider receives one.
Evidence checks G1–G3 cover provider identity, commercial terms and evidence integrity alongside the score.
Dates and version methodology M1.3, score date, and evidence cutoff.

Every published ranking must show PBS, criteria and weights, methodology version, score date, and evidence cutoff.

PBSCriteria and weights Methodology versionScore dateEvidence cutoff

Disclosure

Every provider receives a numeric Provider Buyer Score.

Stages and controls

Stages

Controls

0 The buyer question Every provider is scored

1 Eight criteria B1–B8, fixed weights

2 Incomplete evidence Still scored, never omitted

3 Evidence checks Identity, terms, integrity

Ranking and ties 2-point bands

Version and dates M1.3, cutoff, score date

Stage 0

The buyer question

The ranking answers one question, in full, for every provider we list.

For a typical link-building buyer considering this provider today, how strong is the overall purchase proposition?

Deliverable quality, value, protection, verified customer experience, managed help, commercial flexibility, reporting, and continuity.

Every provider is scored

Every criterion receives a number on a 0–100 scale.
PBS is the weighted sum of those eight numbers.
A provider with sparse public detail still receives a PBS and can be ordered.

The number is a Provider Buyer Score. It is always calculated. A thin public record does not remove the provider from the ranking or leave the score blank.

Stage 1

Eight criteria

PBS scores the provider-level buyer proposition. The eight weights are fixed for M1.3.

PBS = Σᵢ₌₁⁸ wᵢ sᵢ

PBS = sum of (weight × criterion score)

ID	Criterion	What is scored	Weight
B1	Deliverable quality and buyer fit	How well the delivered work matches what a typical buyer is paying for	25%
B2	Value for money	Cost against comparable contracted output	20%
B3	Delivery reliability and buyer protection	Whether the work arrives as promised, and what protection exists if it does not	15%
B4	Customer reviews	Trustpilot rating and verified reviews submitted on this site, each weighted by review count	15%
B5	Managed service, support, and buyer control	Named help, support, and the buyer’s ability to direct the work	10%
B6	Commercial terms and flexibility	Contract terms, cancellation, and room to change scope	7%
B7	Reporting and transparency	Whether reporting shows delivered work clearly and states its limits	5%
B8	Continuity and capacity	Evidence the provider can keep delivering at the stated scope	3%
	Total		100%

The 0–100 scale

Every criterion uses the same numeric scale.

0 evidenced worst state, or none of the measured quality.

25 clearly below what a credible purchase looks like.

50 mixed evidence.

75 strong, consistently evidenced practice.

100 best realistically attainable state, defined through observable evidence.

Each criterion is scored 0–100, then multiplied by its weight. The eight products are added.

Stage 2

Incomplete evidence

Public pages are uneven. M1.3 treats a gap as an unknown detail, not as a failed purchase.

Value for money uses a three-step hierarchy

B2 uses the first method that the public record actually supports, in this order.

A. Comparable basket. Where a matching service-class benchmark exists, compare full minimum-commitment cost per contracted qualifying output with the frozen basket median. Ratio r = comparable unit cost / frozen basket median, interpolated through r=1.75 → 0, 1.30 → 25, 1.00 → 50, 0.80 → 75, 0.65 → 100, and capped at 0/100.
B. Bounded partial comparison. A valid same-class comparison with fewer than three peers may support a directional value score only when scope, minimum commitment, and output basis are materially comparable. The same ratio anchors apply, and the comparability limit is named. One provider’s price is not compared across unlike service classes.
C. No unit comparison. Where no defensible unit comparison exists — including quote-only work, unpublished contracted output counts, or a unique service class — price is not compared against a benchmark. Known price, minimum commitment, and inclusions stay visible. Campaign retainers are not divided by hoped-for earned links.

Refunds, replacements, account management, flexibility, and reporting are scored in B3, B5, B6, and B7. They are not used to pad B2.

The most buyer-favourable current public promise is credited

A provider is expected to honour a current promise it publishes to prospective buyers. When current public pages disagree on a refund, replacement, approval, cancellation, delivery, or support promise, the most buyer-favourable specific version that would reasonably form part of the purchase is scored. The conflict is recorded and does not erase the credited promise unless the versions can be reconciled.

A clearly dated later policy supersedes an older one.
A qualification on the same promise and the same page is part of that promise, not a contradiction to ignore.
Only direct evidence that a promise is withdrawn, inapplicable, or materially false can remove the benefit.
The rule is applied to every provider. The same promise is not counted twice across B3 and B6.

Customer reviews draw on two sources

B4 combines the provider’s public Trustpilot rating with approved reviews submitted on this site by verified customers. Each source is converted to the 0–100 scale from its average rating and weighted by how many reviews it holds, so a handful of ratings moves the score less than a large body of them. Where both sources hold reviews, B4 is their average.

Unless a provider explicitly states otherwise, we treat link-building placements as dofollow. This is our editorial assumption, not a verified link attribute or a provider guarantee. Explicit nofollow, sponsored, mixed-profile and publisher-discretion terms take precedence for the relevant product. Unlinked mentions remain unlinked. The assumption does not establish unique domains, permanence, search rankings or AI visibility.

Stage 3

Evidence checks stay visible

G1 identity and contact, G2 calculability and terms, and G3 evidence integrity stay visible. None of them can suppress a numeric PBS.

Gate	What is checked	Statuspass / hold / fail
G1Identity and contact	A verifiable trading identity and a working way to make contact	pass hold fail
G2Calculability and terms	Enough public commercial detail to understand what is being bought and on what terms	pass hold fail
G3Evidence integrity	No established systematic fabrication, bought positive reviews, self-controlled reviews presented as independent, or material misdescription	pass hold fail

pass

no material unresolved concern found.

hold

a material unresolved concern the buyer should see. It does not change the numeric score except through the directly related evidence.

fail

strong evidence of false identity, systematic fabrication, or established material misdescription. A fail can affect the directly related criterion once; it is never an extra multiplier on the whole score.

The checks cover provider identity, commercial terms and evidence integrity alongside the eight weighted criteria.

A fail is labelled, and the score still publishes

Synthetic providers test the model; they are not market claims.

Provider	Gate	PBS	Published result
Sample Volume Seller	G3 fail: established systematic fabrication of delivery evidence	61.4	PBS 61.4 · G3 fail

Result: the numeric score still publishes alongside the G3 evidence status.

A pass, hold or fail records the state of an evidence check. Every provider still receives a numeric score, and directly related evidence affects only its existing criterion.

Control

Ranking, ties, and independent scores

Every provider receives a PBS and can be ordered.
Show PBS to one decimal place.
A 2.0-point tie band means providers inside the band are not described as meaningfully separated.
The table still keeps a deterministic numeric order inside the band.
Adding, removing, or rescoring one provider changes no other provider’s criterion scores or PBS.

How a 2-point tie band is read

Synthetic providers test the model; they are not market claims.

Provider	PBS
Provider Delta	66.4
Provider Echo	65.1

Gap

1.3

Tie band

2.0 points

Result: the 1.3-point gap is inside the 2.0-point band. Both remain in numeric order; they are not described as meaningfully separated.

Providers are ordered by PBS

Providers are ordered by PBS. A gap of 2.0 points or less is a tie band: the table keeps a deterministic numeric order, but those providers are not described as meaningfully separated. Adding or removing a provider changes nobody else’s score.

Control

Version, score date, and evidence dates

Each ranking states which method produced it, and which evidence it could use.

What every ranking shows

Methodology version

M1.3

SOP version

v1.0

Score date

2026-08-30

Evidence cutoff

2026-08-30

Tie band

2.0 points

Status

locked for the 2026-08-30 publication wave

Versioning

Methodology: M1.3.
Major change: a change to the criteria, weights, B2 hierarchy, promise rule, review rule, gates, the formula, or the tie band.
Minor change: clarification proven not to change any score.
A major change requires every affected provider to be rescored before the new ranking publishes.

Corrections and appeals

Publish a free way to submit evidence and appeal.
Acknowledge within 5 working days; provide a substantive response within 20 working days.
Successful appeals produce a corrected score and dated public correction note.

Conflicts

Disclose any significant provider relationship at the ranking, subject to facts and applicable law.
Paid inclusion, if ever introduced, must not change scores or order and must be visually distinguishable.

What stays fixed in M1.3

The criteria, weights, B2 hierarchy, promise rule, review rule, gates, and tie band stay fixed inside M1.3.
A later method publishes as a new version, with affected providers rescored.

Publication requirements

Every ranking page must visibly provide:

the Provider Buyer Score, to one decimal place;
B1–B8 names and weights;
G1–G3 evidence-check statuses alongside the score;
methodology version M1.3, score date, and evidence cutoff;
the 2.0-point tie band;

Structured data may represent only facts and scores visible on the page.

This is not legal advice; applicability depends on facts and jurisdiction.

Disclosure

Every provider receives a numeric Provider Buyer Score.

The criteria, weights, B2 hierarchy, promise rule, review rule, gates, and tie band stay fixed inside M1.3. A later method publishes as a new version, with affected providers rescored.