Find Glassdoor Interview Questions by Role (2026)

Thirdwatch's Glassdoor Scraper returns Glassdoor interview-question summaries on a pay-per-record basis — company-specific question themes from millions of crowd-sourced candidate reports across 1.4M+ companies. Built for interview-prep platform builders, career-coaching SaaS teams, candidate-prep workflows, and recruiter teams analyzing interview-process effectiveness.

Why scrape Glassdoor for interview research

Interview prep is meaningfully more effective when role- and company-specific. According to Glassdoor's 2024 Site Survey, candidates who prepare with company-specific interview questions report 40%+ higher offer-acceptance confidence and materially better outcomes. The blocker for systematic access: Glassdoor doesn't expose an API, and aggregator platforms (LeetCode, Pramp, Interview Cake) focus on general algorithmic prep rather than company-specific questions.

The job-to-be-done is structured. An interview-prep platform builder wants 1,000+ company question banks for the platform's content layer. A career-coaching SaaS scrapes per-company interview themes for personalised candidate prep. A recruiter team studies their own interview process by mining what candidates report on Glassdoor. A candidate themselves wants every available question for the specific companies they're interviewing at. All reduce to per-company review pulls + theme aggregation + role-level pattern extraction.

How does this compare to the alternatives?

Three options for getting Glassdoor interview-question data:

Paid interview-prep SaaS is curated and high-quality but doesn't surface company-specific real-candidate questions. The Glassdoor Scraper actor page gives you the raw structured-data layer.

How to find Glassdoor interview questions in 4 steps

Step 1: How do I authenticate against Apify?

Sign in at apify.com (free tier, no credit card), open Settings → Integrations, and copy your personal API token. Every example below assumes the token is in APIFY_TOKEN:

export APIFY_TOKEN="apify_api_xxxxxxxxxxxxxxxx"

Step 2: How do I pull Glassdoor reviews scoped to interview content?

Set scrapeType: "reviews" and pass company names; Glassdoor reviews include interview-question summaries in pros/cons themes.

import os, requests, pandas as pd

ACTOR = "thirdwatch~glassdoor-scraper"
TOKEN = os.environ["APIFY_TOKEN"]

PEER_SET = ["google", "meta", "openai", "anthropic", "stripe",
            "airbnb", "amazon", "microsoft", "netflix", "apple"]

resp = requests.post(
    f"https://api.apify.com/v2/acts/{ACTOR}/run-sync-get-dataset-items",
    params={"token": TOKEN},
    json={
        "queries": PEER_SET,
        "scrapeType": "reviews",
        "country": "us",
        "maxResults": 100,
    },
    timeout=3600,
)
df = pd.DataFrame(resp.json())
print(f"{len(df)} review records across {df.company_name.nunique()} companies")

10 companies × 10 review records = ~100 records — pennies on pay-per-record pricing.

Step 3: How do I extract interview-related themes?

Filter pros/cons arrays for interview-related keywords; cluster into question categories.

INTERVIEW_KEYWORDS = ["interview", "coding", "system design",
                      "behavioural", "behavioral", "technical screen",
                      "phone screen", "onsite", "leadership principle"]

def has_interview_signal(themes):
    if not isinstance(themes, list):
        return False
    return any(any(kw in t.lower() for kw in INTERVIEW_KEYWORDS) for t in themes)

def extract_interview_themes(themes):
    if not isinstance(themes, list):
        return []
    return [t for t in themes
            if any(kw in t.lower() for kw in INTERVIEW_KEYWORDS)]

df["interview_pros"] = df.pros.apply(extract_interview_themes)
df["interview_cons"] = df.cons.apply(extract_interview_themes)
df["has_interview_content"] = df.pros.apply(has_interview_signal) | df.cons.apply(has_interview_signal)

interview_df = df[df.has_interview_content].copy()
print(f"{len(interview_df)} records with interview content")
print(interview_df[["company_name", "interview_pros", "interview_cons"]].head(10))

Themes mentioning "leadership principle" surface Amazon's process specifically; "system design" themes are universal across tech companies.

Step 4: How do I build a per-company question-bank library?

Aggregate interview themes by company; export as Markdown for an interview-prep platform.

from collections import Counter

bank = {}
for _, row in interview_df.iterrows():
    company = row.company_name
    if company not in bank:
        bank[company] = Counter()
    for theme in row.interview_pros + row.interview_cons:
        bank[company][theme.lower().strip()] += 1

# Export top themes per company
import pathlib
for company, theme_counts in bank.items():
    out = pathlib.Path(f"interview_bank/{company.replace(' ', '_')}.md")
    out.parent.mkdir(parents=True, exist_ok=True)
    lines = [f"# {company} Interview Questions\n",
              "_Sourced from Glassdoor employee-reported interviews._\n"]
    for theme, count in theme_counts.most_common(20):
        lines.append(f"- {theme} _(mentioned {count}x)_")
    out.write_text("\n".join(lines))

print(f"Exported {len(bank)} company question banks")

The Markdown files are ready to ingest into an interview-prep platform's content layer or a candidate-coaching tool.

Sample output

A single review record (with interview-related cons) looks like this:

{
  "company_name": "Amazon",
  "company_rating": 3.6,
  "pros": ["Strong leadership principles training", "High autonomy"],
  "cons": ["Tough interview process — 5 rounds covering each leadership principle in depth",
           "System design rounds particularly difficult",
           "Behavioural questions tied to leadership principles take significant prep"]
}

Interview-related signals concentrate in the cons array because candidates more frequently report tough interview processes than easy ones. Filter on keyword matches in cons for the strongest signal; supplement with pros for context.

Common pitfalls

Three things go wrong in interview-research pipelines on Glassdoor data. Theme abstraction loss — Glassdoor's pros/cons are clustered themes rather than raw question text, so you get "system design questions" rather than the specific question. For raw question text, fetch individual interview-experience pages by URL after identifying companies via the actor. Recency bias — Glassdoor weights recent reviews higher in clustering, so themes reflect the current interview process rather than historical patterns. For long-running prep platforms, refresh quarterly to keep question banks current. Cross-role contamination — Glassdoor's company-level reviews mix interview themes across roles (engineering, product, sales). For role-specific question banks, additional filtering on role-name keywords in the theme strings is required.

Thirdwatch's actor uses Production-grade anti-bot handling bypass for Glassdoor's Cloudflare protection — production-tested at sustained weekly volumes. The 4096 MB max memory and 3,600-second timeout headroom mean even 100-company batch runs complete cleanly. Pair Glassdoor with our LinkedIn Profile Scraper to identify candidates who interviewed at target companies and reach out for richer first-person interview content. A fourth subtle issue worth flagging: Glassdoor's review-clustering algorithm gives extra weight to recent and high-engagement reviews, which means the same company viewed three months apart can surface meaningfully different "top themes" even when the underlying interview process hasn't changed. For longitudinal interview-prep dashboards, snapshot themes monthly and treat the rolling 90-day union as your stable theme set rather than reading any single snapshot as authoritative. A fifth pattern unique to interview-prep extraction: companies with employee-review counts under ~50 produce noisy theme clusters because Glassdoor's clustering threshold is built for higher-volume employers. For early-stage startups, fall back to raw review text concatenation + LLM-based theme extraction rather than relying on the clustered pros/cons arrays — you'll get better coverage at the cost of an extra LLM-call step in the pipeline. A sixth pitfall: when Glassdoor surfaces interview-question themes for very large employers (Amazon, Google, Meta), the themes blur across dozens of role families, so for role-specific question banks at mega-employers always cross-reference the themed cons with the jobTitle field if present, or fall back to filtering at the company-subdivision level (e.g., "Amazon Web Services" rather than "Amazon" as the query).

Related use cases

• Scrape Glassdoor salaries for compensation benchmarking
• Research company reviews on Glassdoor
• Track Glassdoor rating changes over time
• The complete guide to scraping job boards
• All Thirdwatch use-case guides