Generational Content Testing

import os, time, json, requests

API_KEY = os.environ["MAVERA_API_KEY"]
BASE = "https://app.mavera.io/api/v1"
HEADERS = {"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"}

GENERATIONS = [
    {"name": "Jordan — Gen Z Consumer", "role": "Gen Z (born 1997–2012)",
     "description": "Digital native, 24 years old, lives on TikTok and Instagram. Values authenticity and social proof. Skeptical of corporate messaging, responds to creator-style content. Short attention span for traditional ads.",
     "traits": ["digitally native", "authenticity-driven", "socially conscious", "short-form preference"], "gen": "gen_z"},
    {"name": "Morgan — Millennial Professional", "role": "Millennial (born 1981–1996)",
     "description": "35, mid-career professional. Responds to experiences over things, purpose-driven brands, and data-backed claims. Active on Instagram, LinkedIn, and podcasts. Willing to pay premium for convenience.",
     "traits": ["experience-oriented", "purpose-driven", "research-heavy", "subscription-comfortable"], "gen": "millennial"},
    {"name": "Casey — Gen X Decision-Maker", "role": "Gen X (born 1965–1980)",
     "description": "48, peak earning years. Pragmatic and self-reliant. Responds to straightforward value propositions without hype. Brand-loyal when value is consistent. Email and web-first.",
     "traits": ["pragmatic", "self-reliant", "brand-loyal", "hype-resistant"], "gen": "gen_x"},
    {"name": "Pat — Boomer Evaluator", "role": "Baby Boomer (born 1946–1964)",
     "description": "65, retired or senior advisory role. Values trust, credentials, and track record over novelty. Prefers longer-form content — articles, whitepapers, webinars. Active on Facebook and email.",
     "traits": ["trust-driven", "detail-oriented", "authority-responsive", "long-form preference"], "gen": "boomer"},
]

def create_persona(g):
    resp = requests.post(f"{BASE}/personas", headers=HEADERS, json={
        "name": g["name"], "role": g["role"], "description": g["description"], "traits": g["traits"],
    })
    resp.raise_for_status()
    data = resp.json()
    print(f"  Created: {data['name']} → {data['id']}")
    return data["id"]

gen_personas = {}
for g in GENERATIONS:
    gen_personas[g["gen"]] = {"id": create_persona(g), "name": g["name"]}

CAMPAIGN = """
Campaign: "Own Your Future"
Product: Personal finance app for investing, budgeting, and retirement planning.
Tagline: "Smart money moves start here."
Visual: Split-screen — stressed about finances vs. confident and relaxed.
CTA: "Start your free plan today."
Channels: Instagram Reels, YouTube pre-roll, email nurture, LinkedIn ads.
"""

QUESTIONS = [
    {"question": f"Share your honest first impression:\n\n{CAMPAIGN}", "type": "OPEN_ENDED", "order": 1},
    {"question": "How compelling is the tagline 'Smart money moves start here'? (0-10)", "type": "NPS", "order": 2},
    {"question": "Which channel would most effectively reach you?", "type": "MULTIPLE_CHOICE",
     "options": ["Instagram Reels", "YouTube pre-roll", "Email nurture", "LinkedIn ads", "None"], "order": 3},
    {"question": "What emotion does the split-screen visual evoke?", "type": "OPEN_ENDED", "order": 4},
    {"question": "Would you click 'Start your free plan today'?", "type": "MULTIPLE_CHOICE",
     "options": ["Yes — immediately", "Maybe — need more info", "No — not relevant", "No — CTA too generic"], "order": 5},
    {"question": "What one change would make this more appealing to you?", "type": "OPEN_ENDED", "order": 6},
]

def run_gen_fg(gen_key, persona_id):
    resp = requests.post(f"{BASE}/focus-groups", headers=HEADERS, json={
        "name": f"Generational Test — {gen_key}", "persona_ids": [persona_id],
        "sample_size": 10, "questions": QUESTIONS,
    })
    resp.raise_for_status()
    return resp.json()

def poll_fg(fg_id, timeout_min=10):
    for _ in range(timeout_min * 6):
        resp = requests.get(f"{BASE}/focus-groups/{fg_id}", headers=HEADERS).json()
        if "error" in resp: raise Exception(resp["error"]["message"])
        if resp["status"] == "COMPLETED": return resp
        time.sleep(10)
    raise TimeoutError(f"Focus Group {fg_id} timed out")

# Create all FGs before polling — they run concurrently
fg_jobs = {}
for gen_key, info in gen_personas.items():
    fg = run_gen_fg(gen_key, info["id"])
    fg_jobs[gen_key] = fg["id"]

gen_results = {gk: poll_fg(fid) for gk, fid in fg_jobs.items()}
print("All 4 Focus Groups complete.")

def compare_generations(results_by_gen):
    print("\n" + "=" * 70)
    print("GENERATIONAL COMPARISON REPORT")
    print("=" * 70)

    labels = {"gen_z": "Gen Z", "millennial": "Millennial", "gen_x": "Gen X", "boomer": "Boomer"}
    order = ["gen_z", "millennial", "gen_x", "boomer"]

    for q_idx, q in enumerate(QUESTIONS):
        print(f"\n{'─' * 60}\nQ{q_idx+1}: {q['question'].split(chr(10))[0][:70]}\n{'─' * 60}")
        for gk in order:
            qr = results_by_gen[gk]["results"][q_idx]
            lbl = labels[gk]
            if q["type"] == "NPS":
                print(f"  {lbl:12s} NPS: {qr.get('nps_score', 'N/A')}")
            elif q["type"] == "MULTIPLE_CHOICE":
                counts = qr.get("option_counts", {})
                top = max(counts, key=counts.get) if counts else "N/A"
                print(f"  {lbl:12s} Top: {top}")
            else:
                print(f"  {lbl:12s} {(qr.get('summary', 'N/A'))[:90]}")

compare_generations(gen_results)

from openai import OpenAI

mavera = OpenAI(api_key=API_KEY, base_url="https://app.mavera.io/api/v1")

def generate_recs(results_by_gen):
    labels = {"gen_z": "Gen Z", "millennial": "Millennial", "gen_x": "Gen X", "boomer": "Boomer"}
    parts = []
    for gk in ["gen_z", "millennial", "gen_x", "boomer"]:
        p = f"## {labels[gk]}\n"
        for qr in results_by_gen[gk].get("results", []):
            q_short = qr["question"][:50]
            if qr["type"] == "NPS": p += f"- {q_short}: NPS {qr.get('nps_score')}\n"
            elif qr["type"] == "MULTIPLE_CHOICE": p += f"- {q_short}: {qr.get('option_counts', {})}\n"
            else: p += f"- {q_short}: {qr.get('summary', 'N/A')}\n"
        parts.append(p)

    resp = mavera.responses.create(model="mavera-1", input=[{"role": "user", "content":
        "Analyze these Focus Group results from 4 generational segments.\n\n" + "\n".join(parts) +
        "\n\nProduce: 1) Generation resonance ranking 2) Per-generation content recs "
        "3) A universal version 4) Channel allocation by generation"}])
    return resp.output[0].content[0].text

recs = generate_recs(gen_results)
print("\n=== RECOMMENDATIONS ===\n")
print(recs)