In a young help desk, one senior person spends the first hour of every day doing a job a rule could do: reading new tickets, deciding whose they are, setting priorities, forwarding the odd one to the right group. ServiceDesk Plus has three automation layers built exactly for that hour - and for the quiet follow-up work nobody remembers to do.
Automation in ServiceDesk Plus is not one feature, it is a small family. Business rules inspect every new or updated request and act on it - assign a group, set a priority, notify someone. Technician auto assign then picks the person inside that group, by round robin or by current load. Request templates and reply templates remove the typing. Set up in this order, with a short list of rules you can explain on one page, automation removes the daily sorting work without turning your help desk into a machine nobody understands.
A business rule is a standing instruction: when a request matches these conditions, perform these actions. Conditions can read the fields the requester and the template provide - category, site, department, requester, words in the subject - and actions can assign the group or technician, set priority and status, and fire notifications. This is where an hour of morning triage goes to live. Route network problems to the network group, route anything from the accounts department about a specific application to the person who owns it, and raise the priority automatically when the requester is a defined VIP or the subject carries words like "server down". Two disciplines keep rules healthy: they execute in order, so put specific rules before general ones, and every rule earns its place by firing often - a rule that matched twice last year is documentation, not automation.
Once a ticket lands in the right group, technician auto assign chooses the human. ServiceDesk Plus offers the two classic strategies. Round robin deals tickets in turn - simple, predictable, and blind to the fact that one technician is mid-way through a server rebuild. Load balancing counts open tickets and gives the next one to the least loaded - fairer under real conditions, and the usual choice once a group passes three people. Auto assign respects the structure you built earlier: it works inside the assigned group, and technicians marked on leave are skipped, which is one more reason the leave information has to live in the system and not in a chat message. Review the distribution monthly; if one name absorbs everything, the rule set, not the person, is usually the explanation.
Automation also talks, and here restraint is the whole craft. Keep these: the acknowledgement to the requester with the ticket number; the assignment note to the technician or group; SLA warnings as a breach approaches; and the resolution message with the satisfaction survey. Then go through every other notification the product offers and switch off anything nobody acts on - status-change mails for every tiny edit, copies to managers on all tickets, alerts duplicated on mail and app and SMS at once. Notification spam has a real cost measured in the only currency that matters: the day a genuinely critical alert arrives and is ignored like the previous two hundred. Rare and important beats frequent and muted, every time.
The least glamorous automation saves the most minutes. Request templates were covered in the setup guide - they shape what arrives. Reply templates shape what leaves: prepared, well-written answers for the ten situations your desk repeats weekly - password reset done, parts ordered, access granted, vendor case raised, workaround attached. A technician picks the template, adjusts one line, sends. The requester gets a clear, complete answer in the office's voice instead of a tired one-liner, and new technicians sound experienced from their first week. Write these templates from your own best historical replies; they are already sitting in your closed tickets, waiting to be promoted.
Watch how a request moves from the portal to the right technician in the walkthrough - then map the rules onto your own groups.
Watch Full VideoRound robin is fine for a small group with similar work - it is simple and visibly fair. Once ticket sizes vary a lot, load balancing is kinder, because it counts what each technician is already carrying. Either way, mark leave in the system so the rotation skips absent technicians.
Rules run in a defined order and a ticket can match more than one, so order your rules from the most specific to the most general and re-check after any big change. Keep the list short enough to review quarterly; a rule nobody remembers is a future mystery.
Only the visible kind can - and the visible kind is mostly templates and acknowledgements. Write those in a human voice, with a named technician on the ticket, and automation feels like a well-run desk, not a robot. The sorting and routing happen silently, which is exactly where machines belong.