Ticket volume and response counts tell you almost nothing about whether customers are actually getting good support. These five metrics — First Reply Time, Resolution Time, CSAT, First Contact Resolution, and Escalation Rate — are the ones that actually matter, and here's exactly how to track them.
Stop Tracking the Wrong Things
Most support teams drown in data and starve for insight. Ticket volume, agent login hours, tags applied — these numbers are easy to pull from a dashboard but tell you almost nothing about whether customers are actually getting good help. If your support operation is built around metrics that don't connect to customer outcomes, you're optimizing for the appearance of performance rather than the reality of it.
There are five metrics that genuinely matter. They're not new, but they're consistently misunderstood, poorly tracked, or quietly deprioritized in favor of easier numbers. Here's how to track each one properly and what benchmark you should actually be aiming for.
First Reply Time (FRT)
First Reply Time is the gap between a customer sending a support request and receiving a substantive first response — not an auto-acknowledgment, but an actual reply that shows someone engaged with their issue.
How to track it
Most helpdesks (Gorgias, Zendesk, Freshdesk) calculate FRT automatically. The key is excluding auto-replies and filtering by business hours if your team doesn't operate 24/7. Reporting FRT in calendar hours when you only staff 9–5 will make your numbers look terrible and won't reflect real capacity.
What good looks like
For e-commerce, customers increasingly expect a reply within a few hours. Under 4 hours during business hours is a solid baseline; under 1 hour is where top-performing Shopify brands tend to land. For high-volume periods like BFCM, even 4 hours can feel slow — so build headroom into your workflows before peak season hits.
Resolution Time
Resolution Time measures how long it takes to fully close a ticket from the moment it was opened. It's distinct from FRT — you can reply quickly but still leave a customer waiting days for an actual solution.
How to track it
Resolution Time should be measured from first customer message to the ticket status being marked resolved (not just closed by the agent). Watch out for tickets that get closed prematurely and reopen — those should reset the clock. Segment by issue type: a refund request has a different natural resolution timeline than a technical integration problem.
What good looks like
For standard e-commerce queries (order status, return requests, address changes), under 24 hours is the target. Complex issues might reasonably take 48–72 hours, but anything beyond that should be flagged for review. If your average resolution time keeps climbing, it usually signals either insufficient staffing or broken internal processes — not customer complexity.
Customer Satisfaction Score (CSAT)
CSAT is a post-interaction survey asking customers to rate their support experience, typically on a 1–5 or 1–10 scale. It's the most direct measure of whether customers felt helped — and it's the metric executives most often ask about for good reason.
How to track it
Send CSAT surveys immediately after a ticket closes, while the experience is still fresh. Keep the survey short — one rating question and an optional comment field is enough. The score itself is calculated as the percentage of respondents who selected a positive rating (usually 4 or 5 out of 5). Track response rate alongside the score; a 90% CSAT from 10% of customers isn't very meaningful.
What good looks like
Industry benchmarks for e-commerce CSAT sit around 85–90%. Below 80% is a warning sign. Above 90% consistently suggests your team is genuinely delivering, though it's worth checking whether low-satisfaction customers are even responding — unhappy customers often don't bother with surveys, which can inflate your score artificially.
First Contact Resolution Rate (FCR)
FCR measures the percentage of tickets resolved in a single interaction, without the customer needing to follow up or the agent needing to escalate. It's arguably the most powerful efficiency metric in support because it captures both speed and quality simultaneously.
How to track it
FCR is trickier to measure cleanly. A practical method: flag any ticket where the same customer opens a new ticket on the same issue within 7 days of the original being closed. That original ticket failed FCR. Some teams track it manually through tagging; others use helpdesk automations to detect reopens. Either way, consistency in definition matters more than the specific method.
What good looks like
A strong FCR for e-commerce support is 70–80%. If you're below 60%, customers are routinely coming back with follow-up questions — which means your agents are either not fully resolving issues, or not communicating clearly enough. High FCR directly reduces ticket volume and improves CSAT, making it a force-multiplier metric worth serious attention.
Escalation Rate
Escalation Rate tracks the percentage of tickets that get passed from a frontline agent to a senior agent, manager, or specialist. Some escalations are inevitable — fraud cases, legal questions, complex technical issues — but a high rate often signals that frontline agents lack the information, authority, or training to handle common issues independently.
How to track it
Define escalation clearly: internal reassignments to senior agents, transfers to different departments, or tickets flagged for manager review. Log these consistently through your helpdesk's tagging or assignment workflows. Break escalation data down by issue type and by individual agent — you'll often find patterns that point directly to training gaps or policy gaps.
What good looks like
There's no universal benchmark, but under 10% escalation rate for a well-staffed e-commerce team is reasonable. If escalation is consistently higher, audit what's being escalated. Often it's the same handful of issue types — and solving those at the process level (clearer policies, better agent tools, expanded agent authority) brings the number down quickly.
How These Five Metrics Work Together
None of these metrics tells the full story in isolation. A team can have a great FRT but a terrible FCR — they're fast to reply but never actually solve the problem. A high CSAT with a low FCR might mean customers are happy with friendly responses even when issues drag on, which is a ticking clock on retention. Read them as a system, not individually, and you'll get a much clearer picture of where your support operation is genuinely strong and where it's masking problems.
Set a monthly cadence to review all five together. When one metric moves, check whether the others moved with it. That correlation is where the real diagnostic insight lives.
A Note on Tooling
Tracking these metrics manually is painful and error-prone at scale. Most modern helpdesks surface FRT, Resolution Time, and CSAT natively. FCR and Escalation Rate often require some custom tagging or workflow configuration, but the setup is worth the one-time effort.
Retenza's AI-assisted support layer for Shopify stores is built with these outcome metrics in mind — automatically handling high-FCR ticket types like order status requests and return queries, while routing edge cases to human agents for review. The result is a measurable improvement in both FRT and FCR without expanding headcount, because the right tickets are handled the right way from the first interaction.