Employee referral: the cheapest hiring channel and its blind spot
Employee referrals hire faster and cheaper than any advert. How to run a scheme, what to pay and when, and the homogeneity problem that comes with it.
The difference between a ticket resolved in one reply and one that takes four is almost always the first message. What you ask for at intake decides the whole exchange.
A support ticket is the record of one request from arrival to resolution. Whether it takes one exchange or five is decided almost entirely at intake: a ticket that says it is not working starts a round trip that could have been avoided by asking three questions on the form the customer already had open.
That is the whole design problem. Every field you add improves the tickets you receive and reduces how many people finish the form. The balance is not a matter of taste — ask for what genuinely changes your first reply, and nothing else.
Priority should be set from impact, not from tone. The angriest message is frequently not the most blocked person, and a queue sorted by how forcefully somebody wrote teaches customers that escalation works. Ask for impact explicitly and sort on that; it is fairer and it produces better decisions when everything is urgent at once. Where volume outgrows a shared mailbox, the same fields become the case record in a help desk.
Three are enough for most teams: new, waiting on us, waiting on them. That last one matters more than it looks — it separates our backlog from the tickets stalled on a customer reply, and without it every queue looks worse than it is and nobody knows which number to act on. Add a fourth, resolved, and stop there. Elaborate status schemes with pending, on hold, escalated and monitoring create the illusion of process and quietly become places where tickets go to sit.
Say what you did, not only that it is done. A resolution note the customer can read tells them whether it will happen again and often prevents the next ticket. Where the cause was on your side, say so plainly. Where the same ticket has now arrived three times, the resolution is a documentation or product change rather than a third reply — the pattern-finding half of this is covered in help desk.
Formal complaints deserve their own route. A customer complaint form that asks what happened and what outcome the customer wants makes the complaint actionable from the first contact.
One number tells you whether the queue is really shrinking. First call resolution counts the contacts finished in one interaction, measured by whether the customer comes back within a week rather than by whether the agent pressed close.
Ettex Forms is the intake: a short form anyone can reach without an account, with attachments, landing as structured records in Ettex Records where they can be filtered by impact and status and where the history stays attached to the customer in Ettex CRM.
The boundary, stated plainly: this is not a ticketing system. There is no queue automation, no service level timers, no customer portal where people track their own requests, and no live chat. For a team handling a handful of requests a week the form-and-records approach genuinely works; past that, a dedicated help desk product is the right purchase and we would rather say so. Where response times are promised to a customer, the fields also have to support the service level agreement being measured.
What the person was trying to do, what happened, when it started, and the impact. Five fields or fewer — beyond that, completion drops faster than ticket quality improves.
By impact on the customer, asked for explicitly, rather than by the tone of the message. Sorting by forcefulness teaches people that escalation is how you get served.
New, waiting on us, waiting on them, resolved. The distinction between the two waiting states is what stops the queue looking worse than it is.
It helps at volume and matters less than a prompt acknowledgement with a realistic expectation. Most complaints about support are about silence rather than about speed.
Employee referrals hire faster and cheaper than any advert. How to run a scheme, what to pay and when, and the homogeneity problem that comes with it.
Screening decides who gets a conversation. How to set criteria before reading applications, what to screen on by phone, and the shortcuts that quietly become discrimination.
Conversion tracking records when a visitor does what the page existed for. Which events to define, why platform numbers disagree, and the double-counting nobody checks.