> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mavera.io/llms.txt
> Use this file to discover all available pages before exploring further.
# Campaign Performance → Content Optimization
> Pull top-performing Mailchimp campaign data and extract winning content patterns into a Mavera Brand Voice trained on your actual email performance
### Scenario
Your Mailchimp account has months of campaign data — open rates, click rates, bounce rates, and the actual content that drove those numbers. You pull performance reports for your top-performing campaigns, extract the subject lines, preview text, and content patterns, then feed them into Mavera's Brand Voice to capture what works. The result is a brand voice trained on your actual winning email content.
### Architecture
```mermaid theme={"dark"}
flowchart LR
A["GET /campaigns"] --> B["GET /reports/{id}"] --> C["Extract winning content patterns"] --> D["POST /api/v1/brand-voices"] --> E["Brand voice from winning emails"]
```
### Code
```python Python theme={"dark"}
import os, requests, time
MC_KEY = os.environ["MAILCHIMP_API_KEY"]
MC_DC = os.environ["MAILCHIMP_DC"]
MV = os.environ["MAVERA_API_KEY"]
MB = "https://app.mavera.io/api/v1"
MH = {"Authorization": f"Bearer {MV}", "Content-Type": "application/json"}
MC_BASE = f"https://{MC_DC}.api.mailchimp.com/3.0"
mc_auth = ("anystring", MC_KEY)
r = requests.get(
f"{MC_BASE}/campaigns",
auth=mc_auth,
params={
"count": 50,
"sort_field": "send_time",
"sort_dir": "DESC",
"status": "sent",
"fields": "campaigns.id,campaigns.settings,campaigns.report_summary,campaigns.send_time",
},
)
r.raise_for_status()
campaigns = r.json().get("campaigns", [])
scored = []
for c in campaigns:
report = c.get("report_summary", {})
opens = report.get("open_rate", 0)
clicks = report.get("click_rate", 0)
settings = c.get("settings", {})
if opens > 0:
scored.append({
"id": c["id"],
"subject": settings.get("subject_line", ""),
"preview": settings.get("preview_text", ""),
"from_name": settings.get("from_name", ""),
"open_rate": opens,
"click_rate": clicks,
"send_time": c.get("send_time", ""),
})
scored.sort(key=lambda x: x["open_rate"] * 0.6 + x["click_rate"] * 0.4, reverse=True)
top = scored[:15]
bottom = scored[-5:]
winning_content = []
for camp in top[:10]:
report_r = requests.get(f"{MC_BASE}/reports/{camp['id']}", auth=mc_auth)
if report_r.status_code == 200:
detail = report_r.json()
camp["unique_opens"] = detail.get("opens", {}).get("unique_opens", 0)
camp["unique_clicks"] = detail.get("clicks", {}).get("unique_clicks", 0)
camp["bounces"] = detail.get("bounces", {}).get("hard_bounces", 0)
winning_content.append(
f"Subject: {camp['subject']}\n"
f"Preview: {camp['preview']}\n"
f"From: {camp['from_name']}\n"
f"Open: {camp['open_rate']:.0%} | Click: {camp['click_rate']:.0%}"
)
time.sleep(0.3)
samples_text = "\n\n---\n\n".join(winning_content)
voice = requests.post(f"{MB}/brand-voices", headers=MH, json={
"name": "Mailchimp: Winning Email Voice",
"description": (
"Brand voice extracted from top-performing Mailchimp campaigns. "
f"Based on {len(top)} campaigns with avg open rate "
f"{sum(c['open_rate'] for c in top)/len(top):.0%} and avg click rate "
f"{sum(c['click_rate'] for c in top)/len(top):.0%}."
),
"samples": [c["subject"] + " — " + c["preview"] for c in top],
"guidelines": (
f"Top-performing subject lines average {sum(len(c['subject']) for c in top)//len(top)} characters. "
f"Common patterns: "
+ ("personalization, " if any("{" in c["subject"] or "you" in c["subject"].lower() for c in top) else "")
+ ("urgency, " if any(w in " ".join(c["subject"].lower() for c in top) for w in ["today", "now", "last", "ending"]) else "")
+ ("questions, " if any("?" in c["subject"] for c in top) else "")
+ ("numbers/lists." if any(c["subject"][0].isdigit() for c in top) else "clarity.")
),
}).json()
print(f"Brand Voice created: {voice.get('id')}")
print(f"Based on {len(top)} top campaigns")
print(f"Avg open: {sum(c['open_rate'] for c in top)/len(top):.0%}")
print(f"Avg click: {sum(c['click_rate'] for c in top)/len(top):.0%}")
print(f"\nTop 3 subjects:")
for c in top[:3]:
print(f" {c['open_rate']:.0%} open | {c['subject']}")
```
```javascript JavaScript theme={"dark"}
const MC_KEY = process.env.MAILCHIMP_API_KEY;
const MC_DC = process.env.MAILCHIMP_DC;
const MV = process.env.MAVERA_API_KEY;
const MB = "https://app.mavera.io/api/v1";
const MH = { Authorization: `Bearer ${MV}`, "Content-Type": "application/json" };
const MC_BASE = `https://${MC_DC}.api.mailchimp.com/3.0`;
const mcAuth = "Basic " + Buffer.from(`anystring:${MC_KEY}`).toString("base64");
const campRes = await fetch(
`${MC_BASE}/campaigns?count=50&sort_field=send_time&sort_dir=DESC&status=sent`,
{ headers: { Authorization: mcAuth } }
).then((r) => r.json());
const scored = (campRes.campaigns || [])
.filter((c) => (c.report_summary?.open_rate || 0) > 0)
.map((c) => ({
id: c.id,
subject: c.settings?.subject_line || "",
preview: c.settings?.preview_text || "",
fromName: c.settings?.from_name || "",
openRate: c.report_summary?.open_rate || 0,
clickRate: c.report_summary?.click_rate || 0,
}))
.sort((a, b) => (b.openRate * 0.6 + b.clickRate * 0.4) - (a.openRate * 0.6 + a.clickRate * 0.4));
const top = scored.slice(0, 15);
for (const camp of top.slice(0, 10)) {
const detail = await fetch(`${MC_BASE}/reports/${camp.id}`, {
headers: { Authorization: mcAuth },
}).then((r) => r.json());
camp.uniqueOpens = detail.opens?.unique_opens || 0;
camp.uniqueClicks = detail.clicks?.unique_clicks || 0;
await new Promise((r) => setTimeout(r, 300));
}
const avgOpen = top.reduce((s, c) => s + c.openRate, 0) / top.length;
const avgClick = top.reduce((s, c) => s + c.clickRate, 0) / top.length;
const avgLen = Math.round(top.reduce((s, c) => s + c.subject.length, 0) / top.length);
const voice = await fetch(`${MB}/brand-voices`, { method: "POST", headers: MH,
body: JSON.stringify({
name: "Mailchimp: Winning Email Voice",
description: `Voice from top ${top.length} campaigns. Avg open: ${(avgOpen * 100).toFixed(0)}%. Avg click: ${(avgClick * 100).toFixed(0)}%.`,
samples: top.map((c) => `${c.subject} — ${c.preview}`),
guidelines: `Top subjects avg ${avgLen} chars. Focus on clarity and reader benefit.`,
}),
}).then((r) => r.json());
console.log(`Brand Voice: ${voice.id}`);
console.log(`Based on ${top.length} campaigns | Open: ${(avgOpen * 100).toFixed(0)}% | Click: ${(avgClick * 100).toFixed(0)}%`);
top.slice(0, 3).forEach((c) => console.log(` ${(c.openRate * 100).toFixed(0)}% | ${c.subject}`));
```
### Example Output
```text theme={"dark"}
Brand Voice created: bv_mc_email_1
Based on 15 top campaigns
Avg open: 34%
Avg click: 5.2%
Top 3 subjects:
48% open | Your competitor just launched this — here's your move
44% open | 3 things your dashboard isn't telling you
41% open | We rebuilt the feature you asked for
```
### Error Handling
The `report_summary` field is included in campaign list responses only for sent campaigns. Draft and scheduled campaigns won't have it. Filter by `status=sent`.
Each API call occupies a connection slot. The detailed report fetch for 10 campaigns uses 10 sequential calls with 300ms delays. Avoid parallel fetches exceeding 10.
Mavera's Brand Voice endpoint may have limits on the number of samples. If 15 samples exceeds the limit, reduce to 10 — quality of top performers matters more than quantity.
***
## What's Next
Back to Mailchimp integration overview
Build personas from audience segments
Understand why A/B variants win
Full reference for POST /api/v1/brand-voices