📑 App Review Deep Report — Sentiment, Themes & Complaints

Stop paying $50–100+/month for a review-intelligence subscription you use a few times. This actor delivers the same deep review analysis per report, on demand, with no subscription — one app in, one complete intelligence report out.

	Typical review-intelligence SaaS	App Review Deep Report
Pricing	$50–100+/month, every month	$10.00 / report (FREE tier), pay only when you run
Commitment	Annual/monthly lock-in	None — per report
Break-even	—	Run 5+ reports/month and you're still under one month of most subscriptions

If you pull a competitive or portfolio review read once a week, that's ~$40/month at the FREE tier (less on volume tiers) — versus $50–100+/month for an always-on seat you rarely max out.

Output is one None row per result; billing is pay-per-event, the value event being one deep report (a $0.05 start fee per run, then $10.00 per deep report). Source: itunes.apple.com, play.google.com.

📊 Sample Output

Real rows from run GSygmzhPL2PFM5vlY on build 0.1.14 (2026-09-17), the same input as the Quick start below — every value is as the source published it (emails masked, long text shortened):

analysis_id	data_start	data_end	data_freshness	sentiment_counts	model
20692d17d8d3b63ee7ce	2026-09-16T04:26:40-07:00	2026-09-16T05:46:16-07:00	2026-09-16T05:46:16-07:00	{"negative": 5, "positive": 5}	ngd-lexicon-v1

The run finished with the status message: SUCCEEDED: emitted 1 item(s)

✅ What you get

Each row is flat JSON with these fields (from the dataset schema and the sample run; a field the source does not publish for a given row is null):

• analysis_id — e.g. 20692d17d8d3b63ee7ce
• review_count (number/null) — Number of public reviews analyzed.
• data_start (string/null) — Earliest review date in the window.
• data_end (string/null) — Latest review date in the window.
• data_freshness (string/null) — Most recent review timestamp.
• sentiment_positive_pct (number/null) — Share of positive reviews.
• sentiment_neutral_pct (number/null) — Share of neutral reviews.
• sentiment_negative_pct (number/null) — Share of negative reviews.
• sentiment_counts (object/null) — Raw sentiment label counts.
• average_sentiment_score (number/null) — Mean ngd-lexicon-v1 sentiment score.
• average_rating (number/null) — Mean star rating.
• themes (array/null) — Top theme clusters (term + mentions).
• complaint_clusters (array/null) — Raw complaint phrase clusters.
• trend_buckets (array/null) — Per-month sentiment buckets.
• developer_response_coverage_pct (number/null) — Share of reviews with a developer response.
• examples (array/null) — Sample classified reviews.
• method (string/null) — Analysis method description.
• model (string/null) — Analysis model id (ngd-lexicon-v1).
• generated_at (string/null) — Report generation timestamp.
• partial (boolean/null) — True if the run hit a budget or delivery cap.
• report_id (string/null) — Deterministic report id (hash of the analyzed reviews).
• app_id (string/null) — The app the report covers.
• store (string/null) — apple or google_play.
• country (string/null) — Storefront country code.
• feature_requests (object/null) — Extracted feature requests: count, share, phrases, examples.
• complaint_ranking (array/null) — Ranked complaint clusters with share and example.
• trend (object/null) — Current-vs-prior-period trend with deltas and direction.
• report_markdown (string/null) — Rendered markdown summary of the whole report.
• report_kv_key (string/null) — Key-value store key holding the markdown report.

What you get in one report

Feed it one app (app_id + store) and it fetches up to 1,000 public reviews (logged-out, public data only) and returns one deep report:

• Full sentiment breakdown — positive / neutral / negative split, average sentiment score, average star rating
• Theme clustering — the terms and phrases reviewers actually use, ranked by mentions
• Complaint ranking — negative-review clusters ranked by frequency, each with its share of negatives and a real example quote
• Feature-request extraction — reviews that ask for something, clustered into the most-requested asks
• Trend vs prior period — current window compared against a prior window (explicit dates, or an automatic median split), with sentiment and rating deltas and a worsening/improving/flat call
• Developer-response coverage — how much of the feedback the developer is answering
• Rendered markdown summary — the whole report as a readable brief (a dataset field and a key-value-store record), ready to paste into a deck or ticket

⚙️ Sample inputs

1. Quick start — the Store example (this is what the sample above came from)

{
  "app_id": "284882215",
  "max_reviews": 10,
  "max_themes": 10
}

The sample run charged exactly: 1 × $0.05 apify-actor-start + 1 × $10.00 deep_report = $10.05 on the Free tier — every delivered row was billed.

2. A smaller, narrowed run

{
  "app_id": "284882215",
  "max_reviews": 10,
  "max_themes": 10
}

A bounded run for checking the field shape before scheduling. app_id narrows what the source is asked for.

3. A full-size run

{
  "app_id": "284882215",
  "max_reviews": 10,
  "max_themes": 10
}

The same input run on a schedule; each run is billed only for what it delivers.

🧾 JSON sample record

One real record from run GSygmzhPL2PFM5vlY, exactly as it lands in the dataset (emails masked, long text shortened):

{
  "analysis_id": "20692d17d8d3b63ee7ce",
  "review_count": 10,
  "data_start": "2026-09-16T04:26:40-07:00",
  "data_end": "2026-09-16T05:46:16-07:00",
  "data_freshness": "2026-09-16T05:46:16-07:00",
  "sentiment_positive_pct": 50,
  "sentiment_neutral_pct": 0,
  "sentiment_negative_pct": 50,
  "sentiment_counts": {
    "negative": 5,
    "positive": 5
  },
  "average_sentiment_score": -0.25,
  "average_rating": 2.9,
  "themes": [
    {
      "theme": "violations",
      "mentions": 2
    },
    {
      "theme": "way",
      "mentions": 2
    },
    {
      "theme": "stops",
      "mentions": 2
    },
    {
      "theme": "feed",
      "mentions": 2
    },
    {
      "theme": "facebook",
      "mentions": 2
    },
    {
      "theme": "here",
      "mentions": 2
    },
    {
      "theme": "nos",
      "mentions": 2
    },
    {
      "theme": "many",
      "mentions": 2
    },
    {
      "theme": "health",
      "mentions": 2
    },
    {
      "theme": "depend",
      "mentions": 1
    }
  ],
  "complaint_clusters": [
    {
      "theme": "violations",
      "mentions": 2
    },
    {
      "theme": "way",
      "mentions": 2
    },
    {
      "theme": "stops",
      "mentions": 2
    },
    {
      "theme": "feed",
      "mentions": 2
    },
    {
      "theme": "depend",
      "mentions": 1
    },
    {
      "theme": "much",
      "mentions": 1
    },
    {
      "theme": "which",
      "mentions": 1
    },
    {
      "theme": "always",
      "mentions": 1
    },
    {
      "theme": "intelligent",
      "mentions": 1
    },
    {
      "theme": "does",
      "mentions": 1
    }
  ],
  "trend_buckets": [
    {
      "period": "2026-09",
      "review_count": 10,
      "average_sentiment_score": -0.25
    }
  ],
  "developer_response_coverage_pct": 0,
  "examples": [
    {
      "label": "negative",
      "score": -3,
      "text": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I reported a comment by a woman wishing for more victims in the me"
    },
    {
      "label": "negative",
      "score": -3,
      "text": "It stops at random while scrolling through my timeline, even worse it stops audio to make you hear the in-app sounds, mainly the pling when a post is posted."
    },
    {
      "label": "negative",
      "score": -3,
      "text": "The app has become a feed of algorithm driven garbage"
    },
    {
      "label": "negative",
      "score": -5,
      "text": "The home feed layout is broken.  Left side column is way too wide, mis-aligned, and text is cut off on the right edge.  Same on two iPad Pro I own.  Did the tester only looked at portrait orientation?  This is so bad!"
    },
    {
      "label": "positive",
      "score": 4,
      "text": "I been on facebook sooo long and I’m still here my business is here we appreciate the love! We ain’t never breaking up! Glorilla voice lol!"
    },
    {
      "label": "positive",
      "score": 3,
      "text": "Me gusta facebook por k es una platforms completa y nos acerca a familias y amigos y nos informants atravez de las noticias.  Me gustas mucho gracias."
    },
    {
      "label": "negative",
      "score": -3,
      "text": "Most useless app ever released in mobile industry.🤬🤬🤬🤬🤬"
    },
    {
      "label": "positive",
      "score": 3,
      "text": "Лучшее приложение для нетворкинг"
    },
    {
      "label": "positive",
      "score": 1.5,
      "text": "Too many ads."
    },
    {
      "label": "positive",
      "score": 3,
      "text": "social media twists so many things and causes mental health and physical problems. americas health care is trash."
    }
  ],
  "method": "transparent lexicon + rating prior; deterministic, no external model",
  "model": "ngd-lexicon-v1",
  "generated_at": "2026-09-17T16:55:32.913467+00:00",
  "partial": false,
  "report_id": "20692d17d8d3b63ee7ce",
  "app_id": "284882215",
  "store": "apple",
  "country": "us",
  "feature_requests": {
    "request_count": 0,
    "request_share_pct": 0,
    "top_request_phrases": [],
    "requests": []
  },
  "complaint_ranking": [
    {
      "rank": 1,
      "complaint": "violations",
      "mentions": 2,
      "share_of_negative_pct": 40,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    },
    {
      "rank": 2,
      "complaint": "way",
      "mentions": 2,
      "share_of_negative_pct": 40,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    },
    {
      "rank": 3,
      "complaint": "stops",
      "mentions": 2,
      "share_of_negative_pct": 40,
      "example": "It stops at random while scrolling through my timeline, even worse it stops audio to make you hear the in-app sounds, mainly the pling when a post is posted."
    },
    {
      "rank": 4,
      "complaint": "feed",
      "mentions": 2,
      "share_of_negative_pct": 40,
      "example": "The app has become a feed of algorithm driven garbage"
    },
    {
      "rank": 5,
      "complaint": "depend",
      "mentions": 1,
      "share_of_negative_pct": 20,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    },
    {
      "rank": 6,
      "complaint": "much",
      "mentions": 1,
      "share_of_negative_pct": 20,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    },
    {
      "rank": 7,
      "complaint": "which",
      "mentions": 1,
      "share_of_negative_pct": 20,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    },
    {
      "rank": 8,
      "complaint": "always",
      "mentions": 1,
      "share_of_negative_pct": 20,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    },
    {
      "rank": 9,
      "complaint": "intelligent",
      "mentions": 1,
      "share_of_negative_pct": 20,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    },
    {
      "rank": 10,
      "complaint": "does",
      "mentions": 1,
      "share_of_negative_pct": 20,
      "example": "You depend so much on AI, which is not always intelligent and does not spot violations. It also flags false violations because of its inability yet to distinguish correctly. I repo"
    }
  ],
  "trend": {
    "available": true,
    "mode": "auto_median_split",
    "current_period": {
      "review_count": 5,
      "data_start": "2026-09-16T04:26:40-07:00",
      "data_end": "2026-09-16T05:12:41-07:00",
      "sentiment_positive_pct": 80,
      "sentiment_neutral_pct": 0,
      "sentiment_negative_pct": 20,
      "average_sentiment_score": 1.5,
      "average_rating": 4
    },
    "prior_period": {
      "review_count": 5,
      "data_start": "2026-09-16T05:16:18-07:00",
      "data_end": "2026-09-16T05:46:16-07:00",
      "sentiment_positive_pct": 20,
      "sentiment_neutral_pct": 0,
      "sentiment_negative_pct": 80,
      "average_sentiment_score": -2,
      "average_rating": 1.8
    },
    "deltas": {
      "sentiment_positive_pct": 60,
      "sentiment_neutral_pct": 0,
      "sentiment_negative_pct": -60,
      "average_sentiment_score": 3.5,
      "average_rating": 2.2
    },
    "sentiment_direction": "improving"
  },
  "report_markdown": "# Deep Review Report — 284882215 (apple) - Reviews analyzed: **10** - Window: 2026-09-16T04:26:40-07:00 → 2026-09-16T05:46:16-07:00 - Average rating: **2.9** - Developer response coverage: 0.0% - Method: transparent lexicon + rating prior; ",
  "report_kv_key": "REPORT_20692d17d8d3b63ee7ce"
}

🔧 How it works

Source. The actor reads itunes.apple.com, play.google.com — endpoints: https://itunes.apple.com/lookup, https://play.google.com/store/apps/details. Public pages and feeds only; nothing behind a login.

Transport. Requests go through Apify residential proxy.

Terminal states. A run ends NORMAL, CAPPED (your cap was reached), PARTIAL (something was withheld and the message says what), GENUINE_EMPTY (the source was read and truly had nothing in scope) or BLOCKED (the source refused or changed shape — the run FAILS loud and bills nothing). A zero-row run is never reported as a silent success.

Charging. Each deep report is charged at the moment it is pushed (deep_report); a row that fails to charge is not delivered, so the dataset count always equals the charged count.

Method — deterministic, transparent, model-included

Analysis runs the ngd-lexicon-v1 engine: a transparent lexicon plus a rating prior. It is fully deterministic — the same reviews always produce the same report. No external LLM, no API key to bring, no "AI" black box. The model is included in the price.

Notes

• Public, logged-out data only. No login, no private endpoints.
• Cost-bounded: memory/time caps, a bounded per-page request-attempt budget, and a hard review/output cap.

What is not done. No login, no cookie or CAPTCHA bypass, no private or personal-account data, no browser automation.

⏰ Set it on a schedule

A single run is a snapshot. The value is the feed: open the actor in Apify Console → Schedules → Create new, add this actor with the input you want, and pick a cadence:

• Daily — 0 7 * * * (07:00 UTC): catch changes within a day.
• Weekly — 0 7 * * 1: a Monday digest.

Each run is independent; keep the same input on the schedule so the results stay comparable. Schedules can also POST results to a webhook (Console → Integrations).

💰 Pricing example

Event	Free	Bronze	Silver	Gold
Actor Start (apify-actor-start)	$0.05	$0.05	$0.05	$0.05
Deep report (deep_report)	$10.00	$9.00	$8.00	$6.70

Worked at the live Free-tier price:

• 1 deep reports: $0.05 start + 1 × $10.00 = $10.05
• 25 deep reports: $0.05 start + 25 × $10.00 = $250.05
• 100 deep reports: $0.05 start + 100 × $10.00 = $1,000.05

A run that delivers zero rows charges the $0.05 start fee only. A BLOCKED run (source refused) fails loud and charges no value event. The start fee is charged once per GB of run memory; the default run memory is 1024 MB.

Yield on the sample run: SUCCEEDED: emitted 1 item(s).

⚖️ Legal & ToS

This actor reads public, logged-out pages and feeds published by itunes.apple.com, play.google.com. It collects only what the source publishes to any visitor, identifies itself, and does not access accounts, private data or anything behind authentication. Use the output in line with the source's terms and your local law; the intended use is B2B research and monitoring.

❓ FAQ

Q: Do I need an API key or a login?
A: No. The source (itunes.apple.com, play.google.com) is read logged-out; the input schema has no key field and the actor carries no secrets.

Q: Why did my run return 0 rows?
A: Read the run's status message. GENUINE_EMPTY means the source was read and had nothing in scope for your input; BLOCKED means the source refused and the run failed without billing a value event — retry later or narrow the input. A zero-row run bills the start fee only.

Q: How fresh is the data?
A: Every run reads the source live at run time; nothing is cached between runs. Put it on a schedule for a continuous feed.

Q: What formats can I export?
A: The dataset downloads as JSON, CSV, Excel, XML or RSS from the run's Dataset tab or the Apify API, and any run can push to a webhook or integration.

Q: How is this different from the other premium report actors actors?
A: Same output shape and billing model; this one covers itunes.apple.com, play.google.com. The siblings under Related Actors cover the other sources or slices — run several on one schedule for a combined feed.

Q: Are there rate limits?
A: The actor paces itself against the source; there is no per-buyer limit beyond your Apify plan's concurrency.

🆘 Troubleshooting

• Run FAILED with BLOCKED → the source refused the request or changed its page shape → nothing was billed beyond the start fee; retry after a while, and if it persists open an Issue with the run id.
• Fewer rows than expected → the source had fewer items in scope → widen the input.
• Input validation error on start → a field is outside the schema's allowed values → start from the Quick start block and change one field at a time.
• Run TIMED-OUT → a very wide request on a slow day → raise the run timeout in Run options or narrow the input; what was delivered before the timeout is still in the dataset.

🔗 Related Actors

• App Review Competitive Report — Head-to-Head Comparison — Stop paying $50–100+/month for a review-intelligence subscription to compare your app against the competition
• ASO Audit Report — App Store Listing Health & Keyword Coverage — Mobile growth / ASO agencies auditing a client app's store listing before a keyword or creative push, and PMs benchmarking their listing against cate…
• EDGAR Diligence Report — Per-Issuer SEC Filing Brief — Per-issuer SEC/EDGAR diligence in one run — material 8-K events, filing history, and risk factors as structured findings plus a rendered markdown bri…
• Grant Funding Report — NIH Awards & Grants.gov Window — Research-development offices and grant consultancies sizing a funder's activity in a topic area, and life-science BD teams tracking award concentrati…
• 🏢 About NexGenData — NexGen Watch is NexGenData's fleet of 256 public monitoring and lookup actors built on official sources, pay-per-result. Browse the catalog at apify.com/nexgenwatch.

⭐ Found this useful?

If this actor saved you a manual check, a quick review on the Apify Store helps other teams find it. Feature request or a source that changed? Open it from the Issues tab — every one is read.