Case study · Technical support

L1 and L2 product support for DevOps SaaS

Company-reported case study

Actigy BPO operated a 24/5 product support desk for CloudSprint. Reported L1 resolution moved from 48% to 79% over five months; code changes and production access remained controlled by the DevOps SaaS company.

Actigy BPO confirms the engagement and supporting records. Results apply to the stated scope and measurement windows.

L1 resolution rate

Result

79%

Baseline: 48%

Window: 5 months

Supporting record: Zendesk analytics

Tier 2 backlog

Result

62

Baseline: 340

Window: 5 months

Supporting record: Jira queue report

First response time

Result

28 minutes

Baseline: 3.2 hours

Window: 5 months

Supporting record: Zendesk analytics

Client profile

CloudSprint, a US DevOps SaaS company with 200 staff.

Challenge and scope

The first line lacks documentation for recurring requests. The second line inherits tickets that could be handled with approved diagnostics and a better knowledge base.

Without a repeatable way to gather logs, product versions, and reproduction details, the support team can send the same issue through two or three queues. L2 then spends time confirming the problem before it can work on it, while engineering receives incomplete issue records and related cases remain open.

Delivery workflow

Cover 5,400 monthly tickets across four product modules. Give L1 a troubleshooting playbook, route defined diagnostics to L2, and send engineering reproducible evidence when product changes are needed.

Classify the ticket by product module, impact, and repeatability. Let L1 use approved diagnostic steps and record what succeeded or failed. For L2 or engineering handoffs, capture an approved log reference, timestamp, environment, reproduction steps, expected behavior, and the workaround communicated to the customer.

Keep Jira and Zendesk records linked so the user can get a status update while investigation continues. Sample escalations for reproducibility and adjust the playbook after clusters of similar tickets. Assign a reviewer to new Confluence articles and maintain the source and revision date for each diagnostic step.

Delivery team

Poland, about 10 FTE. 1 technical lead, 2 L2 engineers and 7 L1 associates.

Systems and tools

Zendesk, Jira, Datadog, Confluence.

Client decision rights

The SaaS company owns code changes, production access, and architecture decisions.

Early milestone

At week 5, L1 resolution reaches 61%.

Outcomes

Company-reported results, with the baseline, measurement window and supporting record described for each measure.

Reported results: L1 and L2 product support for DevOps SaaS
MeasureBaselineResultWindowSupporting record
L1 resolution rate48%79%5 monthsZendesk analytics
Tier 2 backlog340625 monthsJira queue report
First response time3.2 hours28 minutes5 monthsZendesk analytics
Knowledge base articles441865 monthsConfluence inventory

Interpreting the results

The L1 resolution rate rises by 31 percentage points while the L2 backlog drops from 340 to 62. Define which tickets were eligible for L1 resolution and review repeat contacts and reopens. Article count measures coverage when each article has an owner and is used in QA-reviewed cases.

Recommendations for new clients

For SaaS teams with overloaded L2 queues, run a pilot on a representative set of tickets and check reopened cases alongside L1 resolution. Confirm support access and engineering handoff before launch.

Begin with products that generate repeatable questions and one module with common L2 escalations. Provide approved diagnostic access, example defects, and engineering SLAs. Ask the delivery team to resolve a routine ticket and prepare a reproducible escalation, then inspect what evidence can be gathered under the actual access rules.

Scope this workflow for your operation

Use the documented workflow, client-held decisions and operational checks to outline a pilot for your own workload.