Resolution Accuracy
Resolution Accuracy measures whether your bot's answers were factually correct and complete. It is the heaviest of the four Inquio Score™ dimensions, worth 40% of the total.
What it measures
Inquio reads each conversation and judges whether the user actually got the answer they came for.
- answers the question that was asked, not a related one
- gives information that matches your product, pricing, and policy data
- includes everything the user needs to act, with no missing step
- resolves the request within the conversation, without the user having to ask again in a different way
- states something factually wrong (a balance, a fee, a deadline, a condition)
- gives a partial answer that leaves the user to work out the rest
- deflects to a general overview, a link, or a phone number instead of answering
- answers a question the user did not ask
Resolution Accuracy is about correctness of the answer. Whether the conversation felt smooth is measured by Dialog Fluency, and whether the bot resolved it without a handoff is Containment Quality. A bot can be fluent, contained, and still wrong.
How it affects your Inquio Score™
| Dimension | Weight |
|---|---|
| Resolution Accuracy | 40% |
| Containment Quality | 25% |
| Dialog Fluency | 20% |
| Safety and Compliance | 15% |
Because Resolution Accuracy carries the largest weight, it is usually the fastest lever on your overall score. A ten-point gain here moves the total by four points; the same gain in Safety and Compliance moves it by 1.5.
Reading your grade
| Score | Grade | What it means |
|---|---|---|
| 90-100 | A | Answers are reliably correct and complete. |
| 80-89 | B | Mostly correct, with a recurring gap in one or two topics. |
| 70-79 | C | Correct often enough to be useful, wrong often enough to erode trust. |
| 60-69 | D | A substantial share of answers are incomplete or incorrect. |
| 0-59 | F | Users cannot rely on what the bot tells them. |
The trend arrow next to the score compares this report card with the previous one for the same bot. See Trends and benchmarks for how the comparison period is chosen.
What usually drives a low score
No connection to live product data. The most common cause. The bot has been given a general knowledge base but not access to balances, rate sheets, contract terms, or fee schedules, so it answers every specific question with a general summary or a redirect. This shows up in Inquio as a critical issue and almost always affects a cluster of related topics at once.
Stale content. The knowledge base was accurate when it was written. Rates changed, a product was retired, a condition was amended. The bot still answers confidently from the old version.
Topic coverage gaps. Users ask about something the bot was never given an answer for. It improvises, and the improvisation is wrong.
Ambiguous intents. Two topics are close enough that the bot routes to the wrong one and answers a different question correctly.
How to improve it
- Open Issues and Recommendations and filter to issues tagged as affecting Resolution Accuracy. Each one shows an estimated point impact.
- Start with the issue that has the highest impact and the largest number of affected conversations, not the one with the scariest label.
- Read five or six of the affected conversations in full. The pattern is usually obvious by the third one, and it is usually narrower than the issue title suggests.
- Fix the data source before rewriting the copy. Most Resolution Accuracy problems are missing connections, not bad phrasing.
- Re-upload a fresh dataset after the change and compare the trend arrow on the next report card.
Where to find it
- Bot Report Card — the Resolution Accuracy tile, with score, grade, and trend. Click the tile for the breakdown.
- Bot Report Card → Benchmark comparison — your score against the average for your industry.
- Issues and Recommendations — the specific conversations behind the score.