Shopify App Opportunity Finder

Shopify App Opportunity Finder turns up to 20 public Shopify App Store apps and 120 merchant reviews by default into evidence-backed product opportunity reports with transparent query planning and optional BYOK AI. It is for Shopify app developers, ecommerce SaaS teams, agencies, and market researchers who want to find weak incumbent workflows and buildable app ideas.

Run Size and Cost Controls

The default run is bounded so users understand the amount of public data being processed:

• Finds up to 20 Shopify apps from search queries or direct URLs.
• Analyzes up to 8 public reviews per app.
• Analyzes up to 120 reviews total.
• Produces up to 10 ranked opportunities.
• Includes Markdown, JSON, CSV, citations, evidence URLs, and app metadata in one report.

Users can reduce runtime by lowering app/review limits or turning review fetching off. Hard caps are 50 apps, 30 reviews per app, and 500 reviews total.

What You Get

• Direct answer to your analysis goal.
• Ranked Shopify app opportunities with score and suggested product.
• Evidence URLs back to public Shopify App Store app/review pages.
• Apps mentioned per opportunity.
• Pain frequency, urgency, matched keywords, risks, and validation steps.
• Transparent search plan showing generated and applied search queries.
• Optional BYOK AI enhancement for evidence-grounded report polish and opportunity framing.
• Markdown, JSON, and CSV outputs.

Typical Use Cases

• Find Shopify app ideas from low-star merchant reviews.
• Identify weak workflows in returns, inventory, subscriptions, reviews, fulfillment, or support apps.
• Monitor competitor review pain before planning product roadmap.
• Build agency service packages around repeated Shopify merchant complaints.
• Feed structured market intelligence into an AI agent or CRM workflow.

Quick Start

npm install
npm run sample
npm run validate:sample
npm run smoke
npm run validate:smoke

Real-data smoke input:

{
  "searchQueries": ["returns", "inventory sync", "subscriptions"],
  "analysisGoal": "Find Shopify app opportunities with repeated merchant pain, weak incumbent reviews, and clear evidence.",
  "market": "Shopify operations apps",
  "maxApps": 20,
  "includeReviews": true,
  "reviewRatings": ["1", "2", "3"],
  "maxReviewsPerApp": 8,
  "maxTotalReviews": 120,
  "aiProvider": "none",
  "aiMode": "off",
  "maxOpportunities": 10
}

Local outputs:

• outputs/smoke/output.json
• outputs/smoke/report.md
• outputs/smoke/report.csv

Main Inputs

• searchQueries: Shopify App Store queries.
• appUrls: direct Shopify App Store app URLs.
• maxApps: maximum apps to analyze.
• includeReviews: fetch public merchant reviews.
• reviewRatings: ratings to fetch, usually low ratings for pain discovery.
• analysisGoal: report and optional AI analysis direction. It can help generate search queries only when explicit searchQueries are empty.
• market: market context for query planning and opportunity framing.
• maxReviewsPerApp, maxTotalReviews: bounded review controls.
• maxOpportunities: maximum ranked opportunities to output.
• minScore: opportunity score threshold.
• includeRawReviews: include raw review rows in the Dataset item.

searchQueries and appUrls control data collection. analysisGoal controls report framing and optional BYOK AI analysis. The output includes questionPlan so users can inspect generated and applied search queries.

AI / BYOK

AI is optional and off by default. The rules-based report always runs without a model key.

Supported BYOK providers:

• DeepSeek
• OpenAI
• OpenRouter
• Anthropic Claude
• Google Gemini
• Custom OpenAI-compatible endpoint

Inputs:

• aiProvider: provider name.
• aiMode: off, polish, cluster, or full.
• aiApiKey: user-provided secret key. Model API costs are billed by the selected AI provider to the key owner.
• aiModel: model name for the selected provider. DeepSeek defaults to deepseek-chat.
• aiBaseUrl: optional override for custom or compatible endpoints.
• maxAiInputChars, maxAiOutputTokens: call-size limits. They reduce risk but are not exact billing caps.

The AI step can improve report wording, opportunity framing, risks, and validation steps. It cannot create opportunities without existing Shopify app/review evidence.

Output

The first Dataset item includes:

• answer: direct answer to the analysis goal.
• summary: app, review, signal, opportunity, and fetch counts.
• questionPlan: generated and applied Shopify search queries plus warnings.
• aiEnhancement: BYOK AI status, provider, mode, model, and usage metadata when available.
• apps[]: analyzed Shopify app metadata.
• opportunities[]: ranked opportunity objects.
• topOpportunities[]: compact agent-friendly opportunity list.
• signals[]: extracted evidence signals.
• citations[]: evidence URLs and quotes.
• markdownReport: full human-readable report.
• csvReport: opportunity export.
• monetization: PPE charge attempt results.

Each opportunity includes title, score, suggestedProduct, targetMerchant, painSummary, evidence[], evidenceUrl, appsMentioned, painFrequency, urgency, keywords, risks, and nextValidationStep.

Cost Model

There are two cost layers:

• Actor pay-per-event fees: charged to the run owner only when configured in Apify Console and a successful report is generated.
• Apify platform usage: for the normal PPE mode, this is included in the event price and is deducted from the Actor owner's revenue/profit calculation. If the Actor owner enables "Pay per event + usage", the run owner pays platform usage separately.
• BYOK AI provider usage: charged by the selected AI provider to the owner of aiApiKey.

Current public setup: platform usage is not passed through separately, so users see simple PPE event pricing. Higher app/review limits can increase platform usage and reduce the Actor owner's margin, but hard caps and the internal fetch budget keep runs bounded. If monthly PPE revenue does not cover platform costs for this Actor, Apify reports that Actor's profit as zero rather than subtracting the loss from other Actors.

Pricing risk controls:

• Current primary event: report_generated at $0.25; owner gross revenue after Apify margin is $0.20 before platform costs.
• Observed cloud platform cost was about $0.0003 for small real-data runs, far below the primary-event owner gross revenue.
• Public default run size is bounded at 20 apps and 120 reviews; hard caps are 50 apps and 500 reviews.
• Internal Shopify fetch budget is 120 seconds so source fetching should not consume the full Apify run timeout.

The Actor is designed for a simple report_generated PPE event. BYOK AI currently does not add a separate Actor PPE event; provider costs are paid by the key owner. The code records charge results in summary.monetization.

Compliance

• Only bounded public Shopify App Store pages are fetched.
• The Actor does not log in, install apps, submit reviews, or bypass access controls.
• Users are responsible for ensuring their use complies with Shopify terms, Apify terms, and applicable laws.
• Use it for product research, market intelligence, and competitor analysis, not spam, harassment, or deceptive republishing of review content.

Apify Deployment

$npx apify-cli@1.6.1 push

Recommended cloud smoke:

$npx apify-cli@1.6.1 call ACTOR_ID --input-file fixtures/smoke-input.json --memory 1024 --timeout 300 --json

Cloud Actor:

• Actor ID: TCZSAa7zwzjIpAZ2n
• Store URL: https://apify.com/signalflow_studio/shopify-app-opportunity-finder
• Latest verified build: pending deployment after BYOK update.
• Default cloud run options: latest build, 1024 MB memory, 600 second timeout.
• Current pricing model: Pay per event with report_generated.

Recent verification:

• Local sample: 3 apps, 15 reviews, 6 opportunities, Markdown/CSV/JSON generated.
• Local real-data smoke: 20 apps, 81 reviews, 254 signals, 7 opportunities, Markdown/CSV/JSON generated.
• Local BYOK DeepSeek smoke: 3 apps, 15 reviews, 6 opportunities, aiEnhancement.enabled: true, evidence-grounded AI report polish applied.
• Local missing-key AI fallback: 2 apps, 4 reviews, 4 opportunities, aiEnhancement.reason: missing_user_aiApiKey, rules-based report still generated.
• Cloud latest pricing/runtime smoke 9aNyAT9DPQ0ox1LCN: build 0.1.6, status SUCCEEDED, 20 apps, 81 reviews, 254 signals, 7 opportunities, REPORT.md generated, report_generated charged once in OUTPUT.json, platform usage cost about $0.0011, default timeout 600 seconds.
• Cloud real-data run qlbwzRJEBrJAfzp3j: build 0.1.2, status SUCCEEDED, 3 apps, 12 reviews, 53 signals, 6 opportunities, REPORT.md generated, report_generated charged once in OUTPUT.json, platform usage cost about $0.0003.
• Cloud latest-build run S5e434a8FMaQelVjN: build 0.1.3, status SUCCEEDED, 2 apps, 6 reviews, 32 signals, 5 opportunities, report_generated charged once in OUTPUT.json, platform usage cost about $0.0003.

Known release note:

• The Actor is public and functional. A custom Actor avatar still needs to be uploaded through Apify Console because the Actor API rejected external pictureUrl values with invalid-picture-url.