Solutions
The same platform, pointed at a different problem.
There is no support edition and no enterprise edition. A solution here is a combination of Epkora products, configured for the job in front of you.
Customer support
Chat lives in one tool, email in another, and the answer to the question lives in a document nobody can find. Agents rebuild context on every request.
Chat and tickets resolve to one record, the customer panel sits beside the thread, and published articles are one click from the composer.
What it takes
One inbox for conversations and tickets, sharing statuses, priorities and categories
Public replies and private notes on the same timeline, so handovers do not lose the reasoning
Articles attached to replies and recorded against the ticket
Built from
Self-service
The same twenty questions arrive every week, and the help centre is either out of date or does not exist yet.
Provision a help centre in a few steps, publish it on your own domain, and let search analytics tell you exactly which answers are missing.
What it takes
Categories, articles, FAQs and news published to your subdomain or custom domain
No-result search rate shows the questions your content does not answer
AMI answers from the same articles before a person is ever involved
Built from
AI-assisted service
Automated answers either send confident nonsense or hand everything to a person the moment it gets interesting.
AMI works inside your workflows with explicit thresholds: a confidence floor on knowledge, a turn limit before handover, and drafts that wait for approval.
What it takes
Knowledge lookups gated on a confidence percentage you choose
AMI hands over after a set number of replies, and the workflow takes the human path
Drafted replies, summaries and suggested priorities that an agent confirms
Built from
Support operations
SLA breaches are discovered after the fact, routing depends on who happens to be looking, and the escalation path is a shared document.
Encode the process as workflows: policies that start the clock, checks that branch on SLA health, and escalations that are recorded rather than remembered.
What it takes
Triggers on approaching and missed SLA targets, not just on breaches
Clocks that pause while you wait on the customer and resume when the ball is back with you
Escalations, breaches and prevented breaches that show up in reporting
Built from
Service leadership
The weekly number comes from a spreadsheet somebody assembles by hand, and it never agrees with what the team is seeing.
Reporting reads the same records the team works in — conversation volume, SLA attainment, CSAT, channel mix, utilisation and coverage.
What it takes
Executive overview with created-versus-resolved volume and channel mix
SLA attainment against per-priority first-response and resolution targets
Capacity and coverage alongside the service numbers, in the same workspace
Built from
Where to start
Most teams start with the conversations.
Live Chat and Support Tickets first
Get every request onto one record. That alone removes the handover cost, and it is what every other product here reads from.
Then knowledge, then automation
Publish the answers you keep repeating, then encode the process around them. AMI has more to work with once both exist.