FAQ

Frequently asked questions and answers

Answers are grouped by service line so you can check the original service-page wording and approval boundaries behind each one.

Explore all service lines

How does organisational memory work?

These questions explain how organisational memory and approved context are handled.

Can conversations and documents become organisational memory?

swarm-brain organises client-approved conversations, decisions, and documents into shared context that role-based AI can carry forward. For each engagement, we agree the access boundary, retention and deletion rules, and the actions people must approve. It does not automatically store public-site material or act without approval.

View the original service answer

What can you hand over for site operations and AI visibility?

These questions explain how site operations and search and AI visibility follow an approval flow.

Does swarm-connect handle local listings or store registration?

swarm-connect checks whether public web and business information is consistent and whether search and answer engines can read the site structure. It is not a map, review, or store-listing agency, and it does not promise visibility. Proposed changes go through preview and approval before they are applied.

View the original service answer

What is GEO, and what does swarm-connect improve?

GEO is the practice of checking and improving content and structure so generative answer engines can understand and cite a website. swarm-connect reviews public site content, search visibility, and observed AI-search results, but it cannot guarantee a particular ranking, mention, or recommendation. Only approved changes are applied from those observations.

View the original service answer

Can I update my site myself or ask swarm-connect to do it?

Site changes follow request → preview → approval → application. Clients who operate directly can manage content within the approved boundary; for delegated work, AI prepares a draft and a person checks it before publication. The actual integrations and permissions are agreed in a meeting.

View the original service answer

What should we check when a new site gets no enquiries?

More visitors alone do not guarantee enquiries. We review the message, contact path, measurement setup, and content that search engines can read so visitors have a clear next action. The first checks depend on the industry and the condition of the existing site.

View the original service answer

Why do site prices vary?

These questions explain how operating scope and conditions shape the price.

Why do website quotes differ so much between providers?

A site quote is not determined by page count alone. Features and integrations, content readiness, operations and security scope, and code and account ownership conditions all change the work. The public prices below are a starting point for comparison; the final scope is confirmed in a meeting.

View the original service answer

Questions to answer before an AI transformation

These questions explain where to start with AI transformation and agent organisation design.

How do we choose work to automate with AI first?

Start with work that repeats and has a reviewable definition of done. We map frequency, inputs, exceptions, and approval points to select a small candidate. Then we diagnose the current flow and set the pilot boundary and human responsibility.

View the original service answer

Why do teams stop using AI tools they bought?

When a purchased AI tool is not used, we first check whether an owner and an approval flow were defined before the tool. Adoption stalls when nobody knows who reviews the result or how to recover from a failure. Consulting defines roles, standards, and operating responsibility before connecting the necessary tools.

View the original service answer

What effect can AI consulting realistically have?

The effect is testable when workflow diagnosis leads to a small pilot and operating standards. We do not promise enterprise-wide automation at once; we choose work to test, define human approval points, and carry the learning into the next step. The size of the effect depends on the work and the organisation.

View the original service answer

Should we hire someone or try AI first?

AI may be worth testing when throughput and consistent rules dominate; hiring may come first when accountability, relationships, or high-risk judgement dominate. We do not apply one answer to every role, and instead examine exceptions, the accountable owner, and the need for face-to-face collaboration. That becomes the boundary between AI and human work.

View the original service answer

How is AI transformation consulting priced?

swarm-consulting does not publish a fixed price. We review the number and complexity of workflows, systems to connect, data and security boundaries, and the pilot and operating-design scope before preparing a tailored estimate. We confirm the scope in a meeting first, then explain the cost.

View the original service answer

What work can AI agents actually take on?

We do not present unverified client stories or performance numbers. We can describe verifiable work Swarmphony applies to its own site, such as approval-led content, search, reporting, and backup flows. Feasibility for another organisation is designed after reviewing its work and access permissions.

View the original service answer

What does AI agent organisation design consulting include?

AI agent organisation design is not just model selection; it sets roles, collaboration, approvals, and failure recovery as an operating structure. We define ownership by workflow, hand-off rules between agents, and the points where people intervene. The rollout is staged around the existing systems and organisation.

View the original service answer