Competitor Content Audit

Audit competitor top pages from SEMrush and validate messaging differentiators with Mavera Focus Groups

Scenario

Understand what content works for competitors — which pages drive the most organic traffic and where messaging gaps exist. Pull url_organic for a competitor, send to Mave for gap analysis, then validate your differentiators with a focus group.

Architecture

Code

Python

import os, requests, csv, io, time

SR, MV = os.environ["SEMRUSH_API_KEY"], os.environ["MAVERA_API_KEY"]
MB = "https://app.mavera.io/api/v1"
MH = {"Authorization": f"Bearer {MV}", "Content-Type": "application/json"}
COMP = "competitor.com"

resp = requests.get("https://api.semrush.com/", params={
    "type": "url_organic", "key": SR, "domain": COMP,
    "database": "us", "display_limit": 50, "display_sort": "tr_desc",
    "export_columns": "Ur,Fl,Fk,Fp,Nq,Td",
})
reader = csv.reader(io.StringIO(resp.text), delimiter=";")
next(reader)
pages = [{"url": r[0], "traffic": int(r[1] or 0), "keywords": int(r[2] or 0),
          "top_kw": r[4]} for r in reader if len(r) >= 6]

page_block = "\n".join(f"- {p['url']} | traffic: {p['traffic']} | top: \"{p['top_kw']}\""
                       for p in sorted(pages, key=lambda p: -p["traffic"])[:25])
mave = requests.post(f"{MB}/mave/chat", headers=MH, json={
    "message": f"Analyze competitor content gaps.\n\nCOMPETITOR: {COMP}\n"
               f"TOP PAGES:\n{page_block}\n\nAnalyze: 1) Content themes 2) Gaps "
               f"3) Messaging patterns 4) Our opportunities 5) 5 article concepts",
}).json()
print(mave.get("content", "")[:1500])

fg = requests.post(f"{MB}/focus-groups", headers=MH, json={
    "name": f"Audit: {COMP}",
    "auto_personas": {"count": 3, "description": "B2B marketing managers evaluating content tools"},
    "questions": [
        "Rank: 1) AI personas from real data 2) Pre-publish focus groups 3) Brand voice extraction",
        "What's missing? What would make you switch tools?",
        f"What would make you trust our content over {COMP}'s?",
    ], "responses_per_persona": 2,
}).json()

for _ in range(24):
    time.sleep(5)
    data = requests.get(f"{MB}/focus-groups/{fg['id']}", headers=MH).json()
    if data.get("status") == "completed": break
for r in data.get("responses", [])[:6]:
    print(f"[{r.get('persona_id','?')}] {r.get('answer','')[:250]}\n")

Example Output

## Themes: SEO guides (38%), tool comparisons (22%), industry reports (18%)

## Opportunities
1. Persona-driven SEO — no competitor covers audience intelligence. We own it.
2. Pre-publish validation — they A/B test after launch. Focus groups pre-launch
   is a unique differentiator.

[per_auto_01] #1: Focus group validation before publishing. Pre-publish testing
  with structured questions is genuinely different.

Error Handling

ERROR 50 :: NOTHING FOUND

The competitor domain may lack organic data in the selected database. Try a different regional database or verify domain spelling.

Traffic estimates are modeled

SEMrush traffic numbers are estimates based on position and CTR models. Use for relative comparison, not absolute counts.

URL encoding

Some URLs in the response may contain encoded characters. Decode before using in further API calls.