Positive scripting means telling the customer what you can and will do, instead of what you can’t. Below are thirty before-and-after reframes, grouped by situation and mapped to the QA criterion each one supports.
In short
- The information usually stays the same. Only the framing changes.
- Customers react to the framing.
- Mapping each reframe to a criterion makes it coachable and scoreable.
- Treat these as patterns and adapt them to your brand voice.
How to use this library
Positive scripting gets a bad name when it turns into a mandatory word list that makes agents sound like a recording. Used well, it is the opposite: a set of patterns an agent internalizes, so that under pressure their default phrasing keeps the customer oriented instead of shutting the conversation down. The information is usually the same either way. What changes is whether the customer hears a door closing or a path forward.
The way to use what follows is to pick the reframes that fit your product and voice, put them into your agent guidance as examples rather than rules, and then, crucially, score for the behavior rather than the exact words. The right-hand column maps each example to the QA criterion it supports, so your scorecard can reward the reframe wherever an agent expresses it in their own words. Scoring the intent rather than the script is what keeps this from becoming the robotic version everyone hates, a point covered in scoring soft skills in QA.
Saying no without a dead end
The single highest-value category. A blunt no ends the conversation; a reframe redirects it to what is possible.
| Situation | Instead of | Try | QA criterion supported |
|---|---|---|---|
| Feature does not exist | We can’t do that. | Here is the closest way to get that result today, and I will pass your request to the team that decides what we build next. | Solution orientation |
| Out of policy | That’s against our policy. | What I can do within our policy is this, and here is why that route protects your account. | Clarity and policy fidelity |
| Request not possible | That’s not possible. | That specific option is not available, but here are two that get you close. | Solution orientation |
| Wrong team | That’s not my department. | The team who owns this is X, and I will connect you with the context already attached so you do not repeat yourself. | Ownership and handoff quality |
| No refund | We don’t give refunds. | A refund is not something I can apply here, and here is the option that does apply to your situation. | Clarity and empathy |
| Cannot expedite | There’s nothing I can do to speed it up. | I cannot change the timeline, but I can set an alert so you hear the moment it moves, and here is exactly when that will be. | Proactivity |
Delays, waiting, and setting expectations
Most frustration with waiting comes from uncertainty, not the wait itself. These reframes replace vague reassurance with a concrete expectation.
| Situation | Instead of | Try | QA criterion supported |
|---|---|---|---|
| Putting on hold | Please hold. | May I take about two minutes to check this properly? I will come right back. | Setting expectations |
| Investigating | I’ll look into it. | I am going to check three things now, and I will tell you what each one shows. | Clarity |
| Unknown timeline | It’ll be done soon. | I expect this by Thursday. If anything changes that, I will tell you before then. | Setting expectations |
| Escalating | I have to escalate this. | I am bringing in a specialist who can resolve this properly, and here is what happens next. | Ownership |
| Following up | We’ll get back to you. | I will personally follow up by Friday, even if it is only to tell you where things stand. | Proactivity and accountability |
| Checking stock or status | Let me see if we have it. | Let me confirm the exact status for you now so you have a real answer, not a guess. | Accuracy |
Errors, apologies, and taking ownership
When something has gone wrong, the reframe moves from a defensive or hollow apology to acknowledgement plus a concrete next step.
| Situation | Instead of | Try | QA criterion supported |
|---|---|---|---|
| Company error | Sorry for any inconvenience. | You are right, this should not have happened, and here is what I am doing to fix it. | Empathy and ownership |
| Repeated issue | I understand your frustration. | This is the second time you have had to raise this, and that is on us. Let me make sure it is the last. | Empathy and accountability |
| Customer misunderstanding | You did it wrong. | It is easy to miss that step, and it is not obvious. Here is the exact spot to change. | Blameless framing |
| Cannot undo | There’s nothing we can do now. | I cannot undo what happened, but here is how I can make the outcome right from here. | Solution orientation |
| Billing mistake | The charge is correct. | Let me walk through the charge line by line so you can see exactly what it covers. | Transparency and accuracy |
| Agent does not know | I don’t know. | I want to give you the right answer rather than a guess, so let me confirm this and come back to you. | Accuracy and honesty |
Handoffs and endings
The last impression is disproportionate. These reframes close the loop rather than trailing off, and they matter as much for an AI agent handing to a human as for a person.
| Situation | Instead of | Try | QA criterion supported |
|---|---|---|---|
| Transferring | I’m transferring you. | I am connecting you to X, who handles this, and I have added everything so far so you will not start over. | Handoff quality |
| Closing | Is there anything else? | Before we finish, is the original issue fully sorted for you? I want to be sure we did not leave a loose end. | Resolution confirmation |
| Customer still unsure | You’re all set. | Here is a short summary of what we did and what to expect next, so you have it in writing. | Clarity and follow-through |
| Follow-up needed | Contact us if it happens again. | If it recurs, reply to this thread and it comes straight back to me with the full history. | Continuity |
| Survey ask | Please rate this chat. | If this was helpful, a quick rating helps me and the team. Either way, thank you for your patience today. | Appropriate tone |
| Ending on a no | There’s nothing more I can do. | I have reached the limit of what I can change here, and I have logged your feedback where it will be seen. Here is what I would do in your position. | Empathy and honesty |
Turning the library into consistent behavior
A phrase library only changes anything if the reframes actually show up in conversations, and stay showing up. That is a QA problem, not a training-day problem. Add the criteria in the right-hand column to your scorecard, phrased as behaviors rather than exact scripts, and score whether the agent redirected, set an expectation, or took ownership, however they worded it.
Doing that consistently across a team means scoring more than a handful of conversations, because positive scripting is exactly the kind of habit that a small sample misses. When you can score every conversation against the reframe criteria, you can see which situations still trigger dead-end language and coach the specific pattern. And because the scores trace back to the exact lines, you can show an agent the moment a reframe would have changed the conversation, rather than handing them a word list and hoping. That is how positive scripting becomes a measurable behavior instead of a poster on the wall.
Frequently asked questions
What is positive scripting in customer service?
Positive scripting is phrasing what an agent can do and will do rather than what they cannot. It replaces dead-end, blaming, or hedging language with wording that keeps the conversation moving and the customer oriented. The information is often identical; only the framing changes, and framing is what the customer reacts to. Done well it is a set of patterns agents internalize, not a rigid word list.
Does positive scripting mean agents sound scripted?
It does when it is enforced as a mandatory word list, which is why teams end up sounding robotic. The fix is to treat the phrases as patterns and to score the behavior rather than the exact words: did the agent redirect to what is possible, set a concrete expectation, take ownership, however they phrased it. Scoring intent rather than script keeps it natural.
How do you turn positive scripting into something you can coach?
Map each reframe to a QA criterion phrased as a behavior, such as solution orientation or setting expectations, and add those criteria to your scorecard. Then score whether agents express the behavior in their own words. Scoring it across every conversation rather than a sample shows which situations still trigger dead-end language, so you can coach the specific pattern with the transcript in front of you.
What situations benefit most from positive scripting?
Saying no, managing delays and waits, apologizing for errors, handing off to another agent, and ending a conversation. These are the moments where blunt or vague language does the most damage, and where a reframe that redirects to what is possible has the biggest effect on how the interaction lands.
Related terms
- Scoring soft skills in QA
- QA scorecard templates
- What is a QA rubric?
- How to score 100% of conversations
- How to turn QA data into coaching
Score your team for the reframes, not the script
Bring a set of conversations and the phrases you wish your team used more often. We will show you how to score positive scripting as a behavior across every conversation, which situations still trigger dead-end language, and how each score traces back to the exact line where a reframe would have changed the outcome.
