At 3am on a Tuesday, a player in Warsaw opens a high-traffic iGaming app and can't complete a deposit. The product itself is fine - the transaction gateway is behaving oddly for a specific Polish payment method. The issue is narrow, documented, and solvable.
Our L1 agent catches it in the queue within 90 seconds. Checks the transaction log. Applies the documented workaround. Closes the ticket. The player completes their deposit. Total resolution time: 4 minutes. The user never knows there was a problem.
That's L1 support working correctly. Everything else in this article is explaining why that outcome doesn't happen by accident.
What L1 support actually is
L1 support - also called tier-1, first-line, or front-line support - is the first layer of your customer-facing support operation. It handles the full incoming volume: account access issues, billing questions, basic troubleshooting, transaction checks, feature questions, status updates, and anything else that can be resolved using a knowledge base and a clear escalation matrix.
Most articles you'll find about L1 support describe internal IT helpdesks - the team that unlocks employee accounts and resets passwords inside a company. That's a valid use of the term. But if you're running a SaaS product, a gaming platform, a fintech app, or any service with real external users, your L1 operation is a different beast: it's customer-facing, it's live, and when it fails, users leave.
L1 is not a junior version of "real" support. It's a distinct function with a distinct scope. The job is not to diagnose unknown problems - it's to resolve known problems as fast as possible and escalate unknown ones to the right tier without delay.
What L1 agents actually do
A well-scoped L1 agent's day looks like this:
- Monitor the queue across active channels - live chat, email ticketing, inbound phone
- Identify issue type on first contact: categorise, don't diagnose
- Apply knowledge base resolution for known issues - this should be 70–80% of total volume
- Check system logs and transaction records for anomalies (critical in fintech and iGaming)
- Escalate to L2 when the issue falls outside the playbook, with structured handoff notes
- Log every interaction with enough detail that anyone picking it up can continue without asking the customer to repeat themselves
Notice what's not on that list: root cause analysis, code-level investigation, backend access, or engineering judgment. Those belong to L2 and L3. The moment your L1 agents are doing those things, you've either misscoped the role or broken your escalation flow.
If your L1 closure rate - the percentage of tickets resolved at L1 without escalation - is below 60%, the problem is almost never the agents. It's the knowledge base, the playbook, or the escalation criteria. Adding headcount to a broken process makes it more expensive, not better.
How the tiers connect
The tier model exists for one reason: to match issue complexity to the cost of the person resolving it. L3 engineering time costs 10–20x more than L1 agent time. Every routine query that reaches an L3 engineer is money spent incorrectly. A well-calibrated tier structure means your most expensive people spend their time on the problems only they can solve.
The 24/7 problem
Most support strategies look reasonable on paper and break the moment you apply time zones to them.
If your users are global - or even just European - they don't stop having problems at 6pm in your local timezone. A deposit failure at 2am that goes unresolved until your team wakes up is not just a missed ticket. For a fintech or iGaming company, it's a chargeback, a complaint, and a user who won't come back.
Building 24/7 L1 coverage in-house requires - at minimum - 4 headcount for a single agent slot: three 8-hour shifts plus leave and overlap coverage. A 2-agent coverage model (which is the practical minimum for queue management) means 8–10 people. Add management, HR, tooling setup, and knowledge base ownership, and you're running a small department for a function that most product companies have never had to think about before.
In-house vs outsourced: the real numbers
| Factor | In-house 24/7 | Outsourced 24/7 |
|---|---|---|
| Headcount needed | 8–10 agents minimum | Managed by provider |
| Monthly cost (all-in) | $15,000–$30,000+ | $7,000–$17,000 |
| Time to operational | 3–6 months | 2–4 weeks |
| Agent attrition overhead | 30–50%/yr - you own it | Provider's responsibility |
| Tooling integration | Your team builds it | Adapts to your stack |
| Management overhead | You hire and manage | Included |
The only scenario where in-house wins is when your product is so sensitive or so proprietary that an external team genuinely cannot be trained on it. For most SaaS and product companies, this constraint doesn't exist - or it's used as a proxy for "we haven't tried."
What a well-run L1 operation looks like in practice
When we set up the L1 operation for Spribe - one of the leaders in the online casino market, best known for Aviator - the configuration looked like this:
- A dedicated L1 team covering the queue 24/7 and running daily operations independently
- Integrated directly into Spribe's existing Jira instance - we work inside the client's tools, not a separate system that needs to be reconciled
- A clear division of responsibility between L1 and L2, with defined criteria for what gets escalated and what gets closed on the first line
- The SLA agreed with the client: first response within 15 minutes, and simple queries resolved within the hour
The result: Spribe's internal product and technical staff field far fewer routine player queries. They see only what genuinely requires their expertise. That's the correct outcome - and it takes a deliberate setup to get there, not just hiring agents and pointing them at a queue.
5 signs your L1 operation is failing
These don't require a formal audit. If three or more are true, you have a process problem:
- Your L2 or engineering team is fielding password resets and billing questions. If specialists are touching routine queries, the escalation boundary is broken.
- First-response time is over 2 hours on any active channel. This is a queue management issue, not a staffing issue.
- Your ticket escalation rate is above 40%. Either your L1 scope is too narrow or your knowledge base doesn't cover what's actually coming in.
- Your knowledge base was last updated more than 60 days ago. Products change. If the KB doesn't, your L1 agents are improvising - and improvisation at scale means inconsistency.
- You have no data on L1 closure rate. If you're not measuring it, you can't manage it. Start there.
Adding headcount to any of these problems makes them more expensive without fixing them. The fix is process, not people.
IMMIDO runs 24/7 L1 operations for iGaming and tech companies across Europe - integrated into your existing tools, managed end-to-end, live in 2–4 weeks. See how we work →