PlatformHow it worksPricingFull Intelligence ReportFor small businessEnterprise Contact
Buyer decision

AI visibility platform
vs agency

This is not really a choice between two products — it is a question of who does which of five jobs: producing evidence, interpreting it, deciding and doing the implementation, carrying accountability for the outcome, and running the re-measurement. A platform is generally strongest at the first and the last. An agency or consultant is generally strongest at interpretation in context and at owning delivery. Most organisations end up combining them, and the useful question is which jobs you can genuinely staff internally.

The free reading costs nothing and commits you to nothing.

The five jobs

Who is typically better at what

Generalisations about typical strengths. Individual agencies and individual platforms vary enormously, and many agencies now run their own measurement tooling.
JobPlatformAgency or consultantInternal team
EvidenceConsistent, repeatable, produced the same way every timeVaries by engagement; often richer on contextDepends entirely on tooling and time available
InterpretationGeneric by design — it does not know your marketUsually the deepest, especially with sector knowledgeRich on internal context, thinner on outside comparison
ImplementationOnly where a managed service exists and access is grantedOften the core offeringFastest where the team owns the systems
AccountabilityFor the measurement being soundContractual, for agreed deliverablesInternal, and hardest to make objective
Re-measurementTrivial — the same method run againDepends on whether the diagnosis is repeatableDepends on discipline more than capability
IndependenceNot selling the fix if measurement is separate from deliveryDiagnosis and delivery often sit with the same partySelf-assessment, with the usual limitations
What a platform is good at

Consistency, comparability and repetition

A platform's real advantage is that it does the same thing every time. The same method applied to your domain last quarter and this one makes the difference between the two attributable to what happened in between. Applied to a competitor, it puts both on one scale so a gap becomes a distance rather than an impression.

The second advantage is that a measurement taking no self-reported input can be handed to a third party — a client, a board, a partner — as something other than an internal opinion. That property is why separating measurement from delivery is worth something even when the same company can do both.

The limitation is equally real: a platform does not know your market, your constraints, your politics or your roadmap. It can tell you an entity is ambiguous; it cannot tell you that the ambiguity exists because two divisions have been arguing about a brand name for three years.

What an agency is good at

Judgement, sequencing and getting things actually done

Capable agencies and consultants do things software structurally cannot. They translate findings into a plan that fits your organisation, negotiate the trade-offs, carry the work through other people's roadmaps, and apply pattern recognition from other clients that no measurement of your domain alone can supply.

They are also, in most engagements, the party that actually ships the change. A finding that nobody implements has produced nothing, and implementation is a delivery discipline rather than an analytical one.

The consideration worth naming is that diagnosis and delivery frequently sit with the same party, which is a conflict rather than a criticism — it exists for OG01 too, since OG01 sells remediation. It is managed by the measurement taking no self-reported input, by findings naming specific things in specific places so an inflated list is checkable, and by the re-measurement using the same method, which means overstating a problem produces an embarrassing second reading rather than a bigger invoice.

Managed remediation

What "managed" does and does not mean

Nothing is changed on your property without you
OG01 can coordinate managed remediation, and it is worth being exact about what that involves. Work proceeds on an agreed scope, with your approval, and only with whatever access you choose to grant. There is no arrangement in which fixes are applied to your site automatically or unattended. Off-site corrections depend on third parties acting on a request and cannot be guaranteed at all.
Managed remediation operates across five Smart Cards — the four scoring cards plus Local Intelligence, which is a remediation workstream and does not contribute to the composite Authority Rating. Many customers take the specification and execute it with their own developer or agency instead; each finding is written to be handed over rather than interpreted, and doing so costs nothing beyond the reading.
Choosing

Which arrangement suits you

  • Self-service if you have a capable technical or SEO team and what is missing is an objective diagnosis to work against. Take the reading, take the roadmap, execute it yourselves, and re-measure. This is the cheapest path and it works.
  • Agency-led if you have no in-house capacity, or the work is entangled with a rebuild, a migration or a rebrand where sequencing and stakeholder management matter more than measurement.
  • Hybrid — the most common outcome. An independent reading sets the baseline and the priorities; your agency or internal team implements; the same reading runs again to check. Splitting diagnosis from delivery keeps both parties honest and gives the agency an external benchmark to be judged against, which good agencies tend to welcome.
  • Managed remediation if the work repeats across many properties, or if it needs capabilities like structured data and entity records that your team does not have and does not want to build.
Limits

What no arrangement can promise

No platform, agency or internal team can guarantee a ranking, a citation or a mention in an AI answer, because the systems producing those answers do not publish how sources are selected and nobody outside them can verify a claimed mechanism. Nor can anyone commit to a timeframe, since recrawling and third-party sources are outside every supplier's control.

One more limit belongs on a page about choosing a supplier. OG01's Authority Rating is produced by OG01 and by nobody else: no AI or search company issues it, endorses it or recognises it, and it describes conditions observed from outside your domain rather than a standing held inside anyone's system. Treat any supplier's score the same way, including this one, and ask what it is actually derived from.

Questions

Decision questions

Do we have to choose between a platform and an agency?
No, and most organisations do not. The common arrangement is an independent measurement setting the baseline and the priorities, with your agency or internal team doing the implementation, then the same measurement again to check. That splits diagnosis from delivery, which is useful for both parties.
Will OG01 make the fixes for us automatically?
No. Managed remediation runs on an agreed scope, with your approval and only with the access you choose to grant — nothing is changed on your property unattended. Off-site corrections additionally depend on third parties acting on a request, which cannot be guaranteed. Plenty of customers never buy remediation at all and hand the specification to their own team.
Our agency already offers AI visibility work. Why add a platform?
Mainly for independence and repeatability. A measurement produced by someone other than the party delivering the work gives both of you a shared benchmark, and running the identical method again afterwards makes the change checkable. It is not a judgement on your agency — good agencies tend to prefer being measured against something external.
Can we just use the free score and do everything ourselves?
Yes. The free reading gives an Authority Rating across the four scoring cards and the highest-priority actions, and for a capable team that is sometimes enough. The paid depths exist because a roadmap you can hand to a developer needs implementation detail per finding, not because the free reading is a teaser.
What does OG01 cost if we only want the measurement?
The initial score is free, and the paid readings are one-off purchases against a single domain rather than a retainer — current amounts are on the pricing page. Only managed remediation, where work is actually being carried out over time, is priced as an ongoing arrangement.
How do we hold anyone accountable for AI visibility work?
By agreeing in advance what will be measured and re-measuring it with an unchanged method. Accountability for an outcome nobody controls is not available from any supplier; accountability for whether specific, named conditions changed is, and it is the only version worth writing into an engagement.
Start with evidence, then decide who does the work.

The free reading commits you to nothing and gives whoever ends up delivering a baseline to be judged against.