跳到正文
DEV Community· sblabs·· 10 天前AI 評分57

如何將 Google Play 與 App Store 評論匯出為統一格式 CSV

How to export Google Play and App Store reviews to CSV (both stores, one format)

AI 導讀

作者介紹如何透過 Apify 上的 App Reviews Scraper,在單次執行中匯出 Google Play 與 App Store 評論,並以相同欄位整理成 CSV。

正文

App reviews are the cheapest user research you'll ever get: thousands of people telling you exactly what's broken and what they want. The problem is getting them out.

  • App Store Connect and Google Play Console only show reviews for your own apps, in two different formats, with clunky exports.
  • Apple's old RSS review feed that many tutorials still point to now returns nothing.
  • Most scraping tools handle one store, so comparing your iOS and Android feedback means stitching two datasets together.

This post shows how to pull reviews for any app (yours or a competitor's) from both stores in one run, with the same fields, ready for a spreadsheet or an LLM.

Disclosure: I built the tool used below. It's pay-per-use on Apify (you pay per review delivered, a fraction of a cent each).

The tool

App Reviews Scraper on Apify. Give it Google Play package names/links and App Store links/IDs, mixed in one list. Each review comes back with:

store, appId, country, rating, title, text, date, appVersion, thumbsUp (helpful votes), developerReply, developerReplyDate, userName, url

It doesn't need a browser or a proxy, so it's fast. In my release test it pulled 4,200 reviews from 7 apps in 2 countries in under 11 seconds.

Want to see the output before writing any code? This one-click example is pre-filled to get the 500 newest Duolingo reviews from both stores. Open it, click Start, and look at the results.

Export reviews from Python

pip install apify-client pandas
from apify_client import ApifyClient
import pandas as pd

client = ApifyClient("YOUR_APIFY_TOKEN")

run = client.actor("sblabs/app-reviews-scraper").call(run_input={
    "apps": [
        "com.duolingo",                                                      # Google Play
        "https://apps.apple.com/us/app/duolingo-language-lessons/id570060128",  # App Store
    ],
    "countries": ["us", "gb"],
    "maxReviewsPerApp": 1000,   # per app, per country
    "sort": "newest",
})

rows = [r for r in client.dataset(run["defaultDatasetId"]).iterate_items() if r["status"] == "ok"]
df = pd.DataFrame(rows)
df.to_csv("duolingo_reviews.csv", index=False)

print(df.groupby("store")["rating"].mean())

You get one CSV with both stores side by side.

Monitor new reviews cheaply

Set sort to newest and pass a date. The tool stops paging as soon as it reaches older reviews, so a daily run only fetches (and charges for) what's new:

run_input = {"apps": ["com.duolingo", "id570060128"], "sort": "newest", "since": "2026-10-01"}

Schedule that daily in Apify (or from cron, n8n, Make) and post the 1-2 star ones to Slack.

Turn reviews into a feature request list with an LLM

low = df[df["rating"] <= 2]["text"].dropna().head(300).tolist()
prompt = (
    "Here are app reviews. Group them into the top 10 recurring problems, "
    "with a count and two short quotes each:\n\n" + "\n---\n".join(low)
)
# send `prompt` to Claude / GPT / your model of choice

Filter by appVersion to see whether a release made things better or worse, and use developerReply to find negative reviews nobody has answered yet.

Notes

  • The App Store keeps a separate review list per country, so add the countries you care about.
  • Apple's feed doesn't include the app version; Google Play's does.
  • Reviewer names are personal data: aggregate and analyse, don't republish.

Questions or a store/country that doesn't work? Open an issue on the Actor page.

來源:DEV Community · dev.to