Form Submission Analysis

Batch-categorize HubSpot form free-text submissions by pain point, urgency, and use case with Mavera Chat

Scenario

Your HubSpot forms have 500+ free-text submissions. You batch them into Mavera Chat with a customer persona and get structured categorization by pain point, urgency, and use case.

Architecture

Code

Python

import os, json, requests
from openai import OpenAI

HS = os.environ["HUBSPOT_ACCESS_TOKEN"]
MV = os.environ["MAVERA_API_KEY"]

# 1. Pull form submissions (paginated)
FORM_ID = "your-form-guid"
subs, after = [], ""
while len(subs) < 200:
    params = {"limit": 50}
    if after: params["after"] = after
    r = requests.get(f"https://api.hubapi.com/form-integrations/v1/submissions/forms/{FORM_ID}",
        headers={"Authorization": f"Bearer {HS}"}, params=params)
    r.raise_for_status()
    data = r.json()
    subs.extend(data.get("results", []))
    after = data.get("paging", {}).get("next", {}).get("after", "")
    if not after: break

# 2. Extract free-text
texts = []
for s in subs:
    for f in s.get("values", []):
        if f.get("name") == "biggest_challenge" and f.get("value", "").strip():
            texts.append(f["value"].strip())

# 3. Categorize in chunks
mavera = OpenAI(api_key=MV, base_url="https://app.mavera.io/api/v1")
schema = {"type": "json_schema", "json_schema": {"name": "cats", "strict": True, "schema": {
    "type": "object", "required": ["categories"],
    "properties": {"categories": {"type": "array", "items": {"type": "object",
        "required": ["index", "pain_point", "urgency", "use_case"],
        "properties": {
            "index": {"type": "number"}, "pain_point": {"type": "string"},
            "urgency": {"type": "string", "enum": ["high", "medium", "low"]},
            "use_case": {"type": "string"},
        }}}}}}}

all_cats = []
for i in range(0, len(texts), 50):
    chunk = texts[i:i+50]
    numbered = "\n".join(f"{j+1}. {t[:200]}" for j, t in enumerate(chunk))
    result = mavera.responses.create(model="mavera-1",
        input=[{"role": "user", "content": f"Categorize these {len(chunk)} form responses by pain point, urgency, use case.\n\n{numbered}"}],
        extra_body={"persona_id": os.environ.get("CUSTOMER_PERSONA_ID",""), "response_format": schema})
    parsed = json.loads(result.output[0].content[0].text)
    all_cats.extend(parsed.get("categories", []))

# 4. Summary
points = {}
for c in all_cats:
    points[c["pain_point"]] = points.get(c["pain_point"], 0) + 1
print(f"Categorized: {len(all_cats)} | Unique pain points: {len(points)}")
for pp, n in sorted(points.items(), key=lambda x: -x[1])[:10]:
    print(f"  {pp}: {n}")
print(f"High urgency: {sum(1 for c in all_cats if c['urgency']=='high')}/{len(all_cats)}")

Example Output

{ "categorized": 487,
  "top_pain_points": [
    { "pain_point": "Manual reporting takes too long", "count": 89 },
    { "pain_point": "No single source of truth", "count": 72 },
    { "pain_point": "Data scattered across tools", "count": 64 }
  ], "urgency": { "high": 142, "medium": 231, "low": 114 } }

Error Handling

Form GUID

Forms API uses a GUID. Find it in HubSpot → Marketing → Forms → URL, or GET /marketing/v3/forms.

Short responses add noise

Filter out responses under 10 characters before sending to Mavera.

---

HubSpot Integration

Chat API