Governed execution. No moral theater and no work behind a black curtain.
We run visibility directions with manual review of sensitive tasks, documented execution, and explicit boundaries. Below is the operating frame, the prohibition list, and the decision logic for sensitive tasks.
What "governed execution" means.
Every work direction has documentation, rationale, an owner, and a reporting trail. Sensitive tasks and tasks in regulated categories go through manual review before they enter the execution queue. Quality control applies to every result before it goes to the client.
Documentation
Every change is recorded: what was done, where, by whom, and for which decision. The execution log is part of the product.
Task logic
Every direction can be traced back to a decision made from an express audit or consultation, or to a trigger from the previous report.
Approval gates
Sensitive tasks pass a five-question admissibility check before they enter work.
Reporting trail
Reporting is not an add-on service. The execution log, observable changes, and next-step logic close every work cycle.
Quality control
Every result passes review before delivery. Control is on our side, not on the side of the partner who did it.
What we do not do.
These are absolute constraints. They stand without caveats — regardless of category, jurisdiction, or commercial pressure.
- We do not use bots or mass automation.
- We do not write fake reviews or publish fake customer testimonials.
- We do not create fake personas or present our participants as someone else.
- We do not run paid advocacy disguised as organic opinion.
- We do not launch coordinated vote or like inflation.
- We do not work against competitors through hidden methods.
- We do not take political or government work — parties, candidates, campaigns, government narratives, geopolitical influence.
- We do not promise guaranteed search positions or guaranteed recommendations from AI systems.
Manual review for sensitive tasks.
Some work directions are not rejected or accepted automatically: regulated categories, jurisdiction-sensitive tasks, tactics that intersect with disclosure rules on specific platforms. Every such task passes five questions before it moves into execution: is it legal in the jurisdiction where the client operates; does it avoid direct deception or identity substitution; can this work be described as a normal market operation; will it avoid destroying trust in the brand or in us if it becomes public; can it be documented as real work with an explainable result.
Programs with real people — logic and limits.
A network of 100,000+ real people, each with their own device. When the product category requires a live human layer, we launch a participant program: real people, briefed, with a disclosed brand relationship, with a documented task and quality control. Not every project uses them — programs are connected when the visibility problem really requires input from a live person. We are talking about disclosed participation and structured contribution — not independent users. Once the brand relationship exists, the participant remains a real person, but for the platform or regulator they are no longer independent.
What we keep in-house, what goes to partners.
Four things stay with us in every project: diagnosis, coordination, quality control, reporting. PR teams, technical vendors, and local executors work inside the project under a shared plan — shared brief, shared rhythm, one operator accountable for the result. If the client already has vendors they trust, we continue working with them. New partners join only when the plan of directions requires capabilities the client does not yet have.
What we do, what we review case by case, what we do not do.
Three work zones. Most work happens in the white zone. Sensitive tasks go through manual review. The prohibited zone is closed — not discussed.
White zone
Clearly permissible work — proceeds without separate case review.
- Audit, diagnosis, and benchmark work, including independent studies when appropriate.
- Work on owned surfaces: site structure, content, FAQ, trust pages.
- Structured data, feeds, machine-readable surfaces, entity alignment.
- Localization and clarity work across multiple languages.
- Knowledge graph cleanup, Wikidata, alignment of the brand's external description.
- Relevant directories, review infrastructure, expert links.
- Preparation of external materials from partners and category experts.
- Interviews, perception research, message comprehension checks.
- Disclosed brand participation in communities and reviews.
- Partner coordination under a shared plan.
Controlled gray zone
Permitted only after manual review and the five-question test.
- Active initiation of conversations in relevant communities where the brand relationship is real and not hidden.
- Sensitive categories: licensed gambling, 18+ content, speculative finance and promotion of crypto projects, medicine with a high level of regulation.
- Jurisdiction-sensitive tasks — we check the client's jurisdiction and registration form separately.
- Structured discussions involving people from our network under our control.
- Campaigns that intersect with disclosure rules on specific platforms — disclosure is mandatory where the platform requires it.
Prohibited zone
Absolute stops. Not discussed — regardless of category, market, or commercial pressure.
- Fake personas, fake accounts, fake expertise.
- Fake reviews, customer testimonials, cases.
- Bots, mass automation, coordinated vote or like inflation.
- Paid advocacy disguised as organic opinion.
- Any form of hidden work against competitors.
- Political and government persuasion work — parties, candidates, campaigns, government narratives, geopolitical influence.
- "Build us any signal we need" offers — refused without discussion.
Gray-zone tasks pass the five-question admissibility test before execution. The test itself is internal; the fact that it exists is public.
If this work format matches how you need to run visibility — let's talk. Talk to a specialist — $200 →