At OttoQA, we have analyzed and scored tens of thousands of customer service calls across every industry you can imagine. Healthcare, collections, financial services, retail, SaaS support, government. This experience has given us something most QA consultants do not have: pattern recognition at scale.
We have seen what separates a QA form that drives real performance improvements from one that checks boxes. We have seen the questions that correlate with CSAT gains and the ones that are just there because someone added them five years ago.
This guide walks you through our complete customer service QA form. Every question. Every listening prompt. The business rationale behind each section. You can copy this form, adapt it for your center, and have a defensible QA program by the end of the week.
Why Your QA Form Matters More Than Your QA Software
A good form with a spreadsheet beats a bad form with a $100,000 platform every time. Software does not fix a flawed scorecard. If your form is measuring the wrong things, you are just measuring the wrong things faster.
A well-designed QA process can identify performance gaps before they become customer problems, provide data-driven coaching that actually changes agent behavior, ensure compliance with regulatory requirements, and create a culture of continuous improvement.
A poorly designed QA process turns agents into script-followers, punishes helpful behavior that deviates from the rubric, and produces scores that nobody trusts or acts on.
We balance the need for efficient service with maintaining high-quality customer interactions. The form integrates essential compliance elements for customer protection while keeping the agent experience human.
Section 1: Call Opening and Greeting
The first 30 seconds of a call set the tone for everything that follows. This section measures whether the agent opened the call professionally and set the customer up for a productive interaction.
a) Did the agent use a proper greeting including company name and their own name? (Yes/No)
What to listen for: "Thank you for calling [Company]. This is [Name]. How can I help you today?"
Why it matters: A proper greeting confirms the customer reached the right company and establishes a personal connection. Skipping the name is a small thing that feels impersonal.
b) Did the agent verify the customer's identity according to security protocols? (Yes/No)
What to listen for: Clear verification questions appropriate for the account sensitivity. For financial or healthcare, this is non-negotiable compliance.
Why it matters: Identity verification is often a regulatory requirement. Skipping it creates legal exposure for the company.
c) Did the agent ask an open-ended question to understand the reason for the call? (Yes/No)
What to listen for: "What can I help you with today?" or "Tell me what is going on." Not "Are you calling about your bill?"
Why it matters: Open-ended questions uncover the real issue faster than yes/no prompts. Closed questions often send the agent down the wrong path.
Section 2: Active Listening and Empathy
This is where most QA forms fall apart. Empathy questions tend to be vague and unmeasurable ("Did the agent show empathy?"). We break empathy into specific, observable behaviors.
a) Did the agent acknowledge the customer's stated concern before moving to resolution? (Yes/No)
What to listen for: "I understand that's frustrating" or "I can see why this would be an issue" BEFORE jumping to fix-it mode.
Why it matters: Customers need to feel heard before they can absorb a solution. Jumping straight to the fix feels dismissive.
b) Did the agent use active listening cues during customer explanations? (Yes/No)
What to listen for: Verbal cues like "I see," "got it," "okay." Or brief restatements: "So what I am hearing is..."
Why it matters: Silence during customer explanations feels like the agent is not paying attention. Active listening signals engagement.
c) Did the agent maintain a calm and composed tone throughout the interaction? (Yes/No)
What to listen for: A consistent, composed tone, even if the customer becomes emotional or aggressive.
Why it matters: A calm agent can de-escalate tense situations and maintain a productive conversation. An agent who matches the customer's agitation makes everything worse.
d) Did the agent avoid interrupting the customer? (Yes/No)
What to listen for: The customer completes their thoughts. If there is interruption, it is only to redirect genuinely off-topic content.
Why it matters: Interrupting signals the agent does not respect the customer or is too focused on their own script.
Section 3: Problem Identification and Solution
This is the core of most support calls. Did the agent actually identify and solve the problem? Or did they follow process while missing the point?
a) Did the agent offer relevant solutions based on the customer's stated issue? (Yes/No)
What to listen for: "Based on what you have told me about [specific issue], I think [specific solution] could help address your concerns."
Why it matters: Tailored solutions demonstrate attentiveness and increase the likelihood of actually resolving the issue.
b) Did the agent effectively communicate the benefits of the proposed solution? (Yes/No)
What to listen for: Clear articulation of how the solution addresses the customer's specific concern. Not just "here is what we will do" but "here is how this helps you."
Why it matters: Customers accept solutions they understand. Process-dumps without context get pushback.
c) Did the agent use appropriate service tactics without being pushy? (Yes/No)
What to look for: Respectful, non-pushy service that builds trust.
Why it matters: Pressure erodes trust and creates long-term damage, even if it wins a short-term outcome.
d) Did the agent attempt to address all aspects of the customer's issue? (Yes/No)
What to check: That no mentioned problems are left unaddressed.
Why it matters: Comprehensive problem-solving prevents follow-up calls and increases first-call resolution rates.
e) Did the agent make at least two attempts to satisfy the customer before escalating? (Yes/No)
What to look for: Multiple, distinct attempts to resolve the issue using different approaches.
Why it matters: Persistence in problem-solving can often resolve issues without the need for escalation, saving time and resources.
Section 4: Product and Process Knowledge
This section measures whether the agent actually knew what they were talking about. Knowledge gaps tank CSAT and drive repeat contacts.
a) Did the agent provide accurate information about products, services, or policies? (Yes/No)
What to verify: Cross-reference the agent's statements against current product documentation or policy.
Why it matters: Wrong information creates downstream issues: billing disputes, follow-up calls, broken trust, regulatory violations.
b) Did the agent use appropriate tools and systems efficiently? (Yes/No)
What to listen for: Minimal dead air while looking up information. No "bear with me while I figure this out" every 30 seconds.
Why it matters: System fluency is a direct proxy for training quality and experience. Slow lookups feel unprofessional.
c) Did the agent ask clarifying questions when needed? (Yes/No)
What to listen for: Appropriate clarifying questions when the customer's issue is ambiguous, rather than assumptions.
Why it matters: Bad assumptions waste time and frustrate customers. Better to ask than to guess.
Section 5: Call Efficiency and Control
Efficiency does not mean rushing. It means not wasting the customer's time.
a) Did the agent keep the call focused on resolving the customer's issue? (Yes/No)
What to listen for: Conversation stays focused on the customer's stated needs, with minimal tangential discussion unless the customer initiates it.
Why it matters: Every minute off-topic is a minute the customer is waiting to get back to their life.
b) Did the agent manage dead air appropriately? (Yes/No)
What to listen for: When lookups take more than 10 seconds, the agent narrates what they are doing: "I am pulling up your account now, give me just a moment."
Why it matters: Unexplained silence makes customers think the call dropped or the agent gave up.
c) Did the agent handle holds professionally? (Yes/No)
What to listen for: Asking permission before placing on hold, giving a time estimate, thanking the customer when returning.
Why it matters: Unannounced holds feel disrespectful. Set expectations and meet them.
Section 6: Closing and Follow-Up
The close is where agents leave their final impression. Rushed closes undo good calls.
a) Did the agent confirm the customer's understanding of the resolution? (Yes/No)
What to listen for: "Just to confirm, here is what we have done..." followed by customer confirmation.
Why it matters: Explicit confirmation catches misunderstandings before they become escalations.
b) Did the agent ask if there was anything else they could help with? (Yes/No)
What to listen for: A sincere "is there anything else I can help you with today?" not a rushed closer.
Why it matters: This single question catches secondary issues that would otherwise become callbacks.
c) Did the agent provide appropriate next steps or follow-up information? (Yes/No)
What to listen for: Clear information about what happens next, expected timelines, and what the customer should do if issues arise.
Why it matters: Clear next steps prevent the "what do I do now" follow-up calls.
Section 7: Compliance and Documentation
This section is industry-specific. For healthcare, financial services, or collections, compliance is pass/fail. One violation can end an agent's career and damage the company.
a) Did the agent follow all required regulatory disclosures? (Yes/No)
What to verify: HIPAA for healthcare, FDCPA/Mini Miranda for collections, PCI-DSS for payments, TCPA for outbound calls.
Why it matters: Compliance failures create legal exposure and potentially massive fines.
b) Did the agent properly document the interaction? (Yes/No)
What to verify: Accurate case notes, correct disposition codes, any required follow-up tasks created.
Why it matters: Bad documentation makes the next agent's job impossible and creates audit problems.
c) Did the agent avoid making promises outside of policy? (Yes/No)
What to listen for: No "I will make sure X happens" unless the agent actually has authority to make X happen.
Why it matters: Over-promised commitments create customer disappointment and damage trust.
Scoring Your QA Form
A few principles for turning these questions into scores:
Use binary yes/no questions, not rating scales. "Did the agent demonstrate empathy" on a 1-5 scale produces inconsistent scores. "Did the agent acknowledge the concern before moving to resolution" is clear and defensible.
Weight compliance questions heavily. A failed compliance question should be able to fail the entire call. A small knowledge gap should not.
Calibrate regularly. Have multiple reviewers score the same calls. Compare scores. Discuss discrepancies. Adjust definitions. Do this monthly.
Connect scoring to coaching. Every failed question should produce a specific coaching action, not just a lower score.
The Auto QA Revolution
While manual QA remains valuable, the landscape is evolving fast. Auto QA uses AI to automatically analyze and score customer interactions across all channels. By implementing Auto QA with OttoQA, you can score 100% of calls instead of the 2-5% you get with manual scoring.
Critically, AI QA only works with a well-built form. Garbage in, garbage out. Take the time to build your scorecard properly using this guide. Then automate the scoring. Both halves matter.
If you want a copy of this form as a starting point for your own center, it is yours. Take it, adapt it, use it. If you want us to score your calls against it automatically, we offer a free 200-call proof of concept to let you see the results before you commit to anything.