A helpdesk that keeps your service commitments

The ticketing system, feature by feature

Full lifecycle. Creation, editing, assignment, statuses with a history of every change, a recycle bin with restore and purge. A ticket can be cloned, split in two when one request was hiding two, or created from a template.

Categorisation on your terms. Categories, ticket types, templates, custom fields and configurable numbering: the form your client sees looks like your trade, not like the vendor's.

Service levels measured, not advertised. Configurable SLA policies, with three distinct targets per priority — acknowledgement, intervention, resolution. Deadlines are calculated, and a regular sweep surfaces breaches before the client reports them.

Traceable collaboration. Public or internal comments, watchers, a complete timeline, a log of the emails exchanged, and suggestions of similar tickets so the same thing is not handled twice.

From time spent to invoice. A timer that starts and stops on the ticket, time and cost entry with an on-site / remote distinction. Billed time is rounded automatically: remotely, a first non-reducible 30-minute block then 15-minute blocks; on site, by half-day started. Each intervention can produce a PDF report carrying your legal identity — company name, legal form, address, registration numbers. The template and the logo of that document are not customisable yet.

Tasks and approvals. Checklists with templates applicable to a ticket, and an approval workflow when an intervention must be signed off before it starts.

Attachments up to 25 MB per file, on the technician side as well as the client side.

Satisfaction. One CSAT survey per ticket, aggregated into the reports.

Knowledge base. Links to your articles from a ticket, suggestions of relevant articles, and Google Drive links when the document already lives there.

Remote control. Start a RustDesk session from the ticket, logged — the trace exists, which is what matters for GDPR. Requires RustDesk configured in your integrations.

Bulk actions to handle a series of tickets in one move.

AI assistance — you connect your own provider. Draft replies, classification into category and priority, sentiment analysis, summaries, duplicate detection. The plumbing is there and accepts several providers (OpenAI, Anthropic, Gemini, or a self-hosted engine): you enable yours, with your key and at your cost. With no provider configured, these features return a deterministic template, labelled as such — useful, but it is not AI and the product does not pretend otherwise.

Steering, and seeing it coming

A dashboard that does not contradict itself. Open tickets, breakdown by status and by priority, SLA breaches, unassigned — all drawn from the same counters as the ticket list. No figure that differs from one screen to the next.

Reports that also state their margin of error. Volume, SLA compliance, mean and median times, overall CSAT, breakdown by category and by type. SLA compliance comes with the size of the sample it was measured on — a rate over three tickets does not mean what a rate over three hundred means.

Per technician, per client, over time. The same indicators broken down by technician, by account, and as time series.

Churn-risk signals. The product cross-references low satisfaction, missed SLAs and acknowledgement times to surface the clients worth calling back — with an acknowledgement of handling, so the signal does not go nowhere.

Saved views. Your filters, kept for yourself or shared with the team.

The client portal, self-service

Your clients open and track their tickets using your request types and your fields, see the history of their company and not only their own, consult linked interventions and attach their files.

They also find your knowledge base there (with voting on how useful articles are), your announcements, the inventory of their equipment and the status of their backups.

A search assistant helps them find the right article before opening a ticket, and pre-fills their request if they open one anyway. It is a search across your articles, with no artificial intelligence — so it always works, depends on no provider, and never creates a ticket on the client's behalf.

The portal carries your brand. Each client is partitioned at database level — they cannot see another client's scope, even by crafting a URL. And access can additionally require two-factor authentication.