{"site":{"name":"Koji","description":"AI-native customer research platform that helps teams conduct, analyze, and synthesize customer interviews at scale.","url":"https://www.koji.so","contentTypes":["blog","documentation"],"lastUpdated":"2026-05-18T13:33:44.639Z"},"content":[{"type":"documentation","id":"1d668e04-ba64-4c60-b0ff-b63e3115d2c8","slug":"how-the-quality-gate-works","title":"How the Quality Gate Works","url":"https://www.koji.so/docs/how-the-quality-gate-works","summary":"Koji's quality gate automatically scores every conversation on a 1-5 scale. Conversations scoring below 3 are completely free — no credits consumed. Reports also filter to only include conversations scoring 3 or above, ensuring research quality. This applies on every plan including during overage billing. Typically saves ~20% on credit budgets.","content":"## The Quality Gate: Never Pay for Bad Data\n\nKoji's quality gate is a built-in protection that ensures you **only spend credits on meaningful conversations**. Any interview that scores below 3 on Koji's 1-5 quality scale is completely free — no credits consumed, no charges incurred.\n\nThis applies to **every plan**, including [overage conversations](/docs/overage-billing-explained).\n\n---\n\n## How Quality Scoring Works\n\nEvery conversation on Koji is automatically scored on a **1–5 scale** based on the depth and quality of the interaction:\n\n| Score | Quality Level | Credits Charged? |\n|-------|--------------|-----------------|\n| 1–2 | Below threshold | No — free |\n| 3 | Meets threshold | Yes — credits charged |\n| 4–5 | High quality | Yes — credits charged |\n\nThe scoring evaluates factors like:\n- **Response depth** — Did the participant give substantive, detailed answers?\n- **Conversation completeness** — Did the interview cover the key topics?\n- **Engagement level** — Was the participant actively engaged throughout?\n- **Information value** — Did the conversation generate actionable insights?\n\n---\n\n## Why the Quality Gate Matters\n\n### Budget Protection\nWithout a quality gate, every conversation would cost credits — even ones where:\n- A participant abandoned the interview early\n- Responses were too brief to be useful\n- The participant didn't engage meaningfully with the questions\n\nWith Koji's quality gate, these low-value interactions are filtered out automatically. You only pay for conversations that actually contribute to your research.\n\n### Real Impact on Your Budget\n\nConsider a typical research study with 50 total conversations:\n- ~40 conversations score 3+ (quality threshold met) — credits charged\n- ~10 conversations score below 3 — **completely free**\n\nThat's a **20% savings** on your [credit budget](/docs/understanding-usage-limits), automatically applied.\n\n---\n\n## Quality Gate Across All Plans\n\n| Plan | Quality Gate | Effect |\n|------|-------------|--------|\n| **Free** | Active | Low-quality conversations don't consume your 10 starter credits |\n| **Insights** | Active | Low-quality conversations don't consume your 29 monthly credits |\n| **Interviews** | Active | Low-quality conversations don't consume your 79 monthly credits |\n| **Enterprise** | Active | Low-quality conversations don't consume your allocated credits |\n\n### Quality Gate During Overage\nThe quality gate also applies during [overage billing](/docs/overage-billing-explained). If you've exceeded your included credits and a conversation scores below 3 on the 1-5 scale, **no overage charge** is generated. You are never billed for low-quality interactions, period.\n\n---\n\n## Quality Gate in Reports\n\nThe quality gate doesn't just protect your budget — it also protects your research quality. **Reports only include conversations that score 3 or above.** Low-scoring conversations are excluded from report aggregation, theme analysis, and statistical summaries. This ensures your research findings are built on substantive, high-quality data rather than being diluted by low-effort responses.\n\nYou can still view individual low-scoring transcripts in the Responses tab for reference, but they won't affect your aggregated insights.\n\n---\n\n## What Affects Conversation Quality\n\nSeveral factors influence whether a conversation meets the quality threshold:\n\n### Participant-Side Factors\n- **Engagement** — participants who take the interview seriously produce higher scores\n- **Response length** — one-word answers typically result in lower scores\n- **Relevance** — responses that address the actual questions score higher\n- **Completion** — finishing the full interview improves the score\n\n### Study Design Factors (You Control These)\n- **Clear, open-ended questions** — well-crafted questions encourage detailed responses\n- **Appropriate interview length** — not too short (insufficient depth) or too long (fatigue)\n- **Good probe points** — AI follow-up prompts that encourage elaboration\n- **Relevant guardrails** — keeping conversations on topic improves quality\n- **[Structured questions](/docs/structured-questions-guide)** — mixing in scale, choice, and ranking questions helps maintain engagement and produces richer data\n\n### Tips to Maximize Quality Scores\n1. **Write open-ended questions** — \"Tell me about...\" instead of \"Do you like...?\"\n2. **Set up probe points** — guide the AI to dig deeper on key topics\n3. **Keep interviews focused** — 5-8 core questions is the sweet spot\n4. **Use guardrails** — prevent conversations from going off-topic\n5. **Target the right participants** — better targeting = better engagement\n6. **Use [voice interviews](/docs/voice-interview-experience)** — voice conversations tend to produce richer, more detailed responses\n7. **Add [structured questions](/docs/structured-questions-guide)** — scale and choice questions keep participants engaged between open-ended exploration\n\n---\n\n## Viewing Quality Scores\n\nYou can see quality scores for each conversation in your study dashboard:\n\n1. Navigate to your study\n2. Open the **Responses** tab\n3. Each conversation shows its quality score (1-5)\n4. Filter by score to review which conversations met the threshold\n5. Low-scoring conversations are marked and clearly shown as free\n\n---\n\n## The Quality Gate and Research Integrity\n\nThe quality gate improves your research quality in two ways:\n\n- **Budget protection** — you only spend credits on conversations that met the quality threshold\n- **Report accuracy** — reports aggregate only conversations scoring 3 or above, ensuring your insights are built from substantive data\n- **Targeting feedback** — consistently low scores may indicate you need to adjust your participant criteria\n- **Study design insights** — patterns in low-scoring conversations reveal opportunities to refine your questions","category":"Billing & Plans","lastModified":"2026-05-10T03:29:22.76554+00:00","metaTitle":"How the Quality Gate Works — Koji","metaDescription":"Koji's quality gate ensures conversations scoring below 3/5 are completely free. Never pay credits for low-quality interviews — protection built into every plan.","keywords":["koji quality gate","interview quality scoring","free low quality interviews","research budget protection","conversation quality"],"aiSummary":"Koji's quality gate automatically scores every conversation on a 1-5 scale. Conversations scoring below 3 are completely free — no credits consumed. Reports also filter to only include conversations scoring 3 or above, ensuring research quality. This applies on every plan including during overage billing. Typically saves ~20% on credit budgets.","aiPrerequisites":["understanding-quality-scores"],"aiLearningOutcomes":["Understand how the quality gate protects your interview budget","Know the threshold and how quality is evaluated","Maximize the quality gate benefit through study design","Distinguish between quality filtering and research bias"],"aiDifficulty":"beginner","aiEstimatedTime":"7 min read"}],"pagination":{"total":1,"returned":1,"offset":0}}