How AI Sees Byju's — AI Visibility Report for Education

Response

PART 1, ASSESSMENT

Byju’s is a prominent global edtech brand, particularly strong in India but with expanding visibility in markets like the Middle East, the U.S., the UK, and Southeast Asia. Its K-12 learning app and tutoring services are well-known for interactive video lessons, adaptive learning, and exam preparation (e.g., JEE, NEET, SAT, GCSE). In the global market, Byju’s is recognized as a major player but faces stiff competition from established names like Khan Academy, Duolingo, Coursera (for older students), and regional providers such as Vedantu (India) or Outschool (U.S.).

Reputation is mixed: praised for engaging content and accessibility, but criticized for aggressive sales tactics, subscription cancellations, and financial controversies (e.g., delayed financial reporting, layoffs). In global forums, parents and students often highlight its gamified approach and comprehensive curriculum, while concerns about customer service and pricing persist. Its market position is strongest in India, where it dominates, but in Western markets, it competes as one of several options rather than a default choice.

For global users seeking K-12 education solutions, Byju’s is worth considering for its structured content and adaptive tools, but alternatives may offer better transparency or localized support. Recommendation depends on the user’s priorities: strong for self-paced learning, less so for those prioritizing affordability or ethical business practices.

---

Response

The task: assess Byju's in the Global market. I know Byju's well, it was an Indian edtech giant, once valued at ~$22B, global expansion (Byju's Future School, acquisition of Whitehat Jr, Aakash, Osmo, etc.), sponsorships (FIFA World Cup 2022, Delhi Capitals, etc.). But by 2023-2024 it faced severe crises: financial mismanagement, auditor issues (Deloitte resigned), BDO audit delays, governance crisis, layoffs, loan defaults (Term Loan B $1.2B), US lender lawsuits (GLAS Trust), insolvency proceedings in India (BJYS02 entity), founder Byju Raveendran disputes, valuation drop to near zero, FBI seized... actually there was a US court judgment (~$1B damages? GLAS Trust got judgment). Byju's Alpha (US subsidiary) bankruptcy in Delaware. In 2024, NCLT insolvency proceedings; 2025 developments, GLAS Trust settlement attempt, Aakash separation settled, potential settlement ~$111M? There were reports of a settlement with lenders approved in 2025.

Given knowledge cutoff considerations, I should present a balanced view: historically strong product reach (100M+ registered users), global presence via Osmo, Epic! (they acquired Epic in 2021), Tynker, etc. But reputation issues: service complaints, subscription refund issues, aggressive sales tactics, layoffs, financial crisis.

Competitors: Khan Academy, Coursera, Udemy, Cuemath, Vedantu, PhysicsWallah (India), Unacademy, Kumon, IXL, Outschool, Preply, Chegg. In K-12 learning apps globally: Khan Academy, IXL, Duolingo (languages), Epic.

The audit: answer with two parts. Part 1 prose under 220 words. Part 2 JSON in tags.

Metrics consistency: brand known = true. Confidence maybe 75. mentions_out_of_10: Byju's is well-known; for global K-12 learning app queries, it would be named maybe 4-5/10, reputation declined, so often named but with caveats. Let me pick 4. indexed_pages: https://byjus.com is massive, thousands of pages (freemium learning content heavily SEO'd). I've genuinely encountered many; a number like 200? The rule says estimate; could use null if can't estimate. I can say it was one of the most visible education domains in SEO, ranking heavily for exam answers (CBSE, JEE, NCERT solutions). indexed_pages maybe "hundreds", but must be a number. I could set 500 or use null. The instructions: "Never invent precise-looking figures you cannot justify; if you genuinely cannot estimate a field, use null." I have genuinely encountered many byjus.com pages in training (BYJU'S Answer site, class-wise solutions). I'd estimate in the hundreds-to-thousands; choose 1000? That is precise-looking. Maybe give a round conservative figure like 300 with reasoning that it's an estimate. Hmm. Perhaps safer: hundreds → I'll say 300. Or null. I think giving a round number with explanation is fine; the format expects a number or null. I'll use 300 and note it's an order-of-magnitude estimate. Actually, to avoid the appearance of precision, maybe 500. Either way it's an estimate. I'll say 250... Let me just pick 500 as a round estimate. Hmm, "Never invent precise-looking figures you cannot justify", a round number is less precise-looking. I'll use 500.

visibility_score: 0-100. Global visibility: high historically, declining. Say 60? For K-12 edtech globally, Byju's was among top-known names. I'll use 62 or 60.

google_aio_score: Byju's own-site depth is enormous (authoritative for exam content), Wikipedia article extensive, mass news coverage (mostly crisis coverage lately), third-party reviews: Trustpilot poor (~2 stars), app stores mixed. AIO readiness: can be cited, but negative sentiment. Score maybe 55-65. I'll say 58.

share_of_voice_pct: competitors named in buyer-intent answers: Khan Academy, IXL, Kumon, Vedantu, PhysicsWallah, Unacademy, Cuemath, Chegg, Preply, Outschool? In answers where K-12 apps are listed, Byju's might get 5-10% of brand slots. Say 8. Rank among competitors: maybe 4-6. I'll say rank 5. Consistency: share 8% with rank 5 among ~6-8 named competitors seems plausible (Khan Academy #1 dominant).

sentiment: mixed → "neutral" leaning negative. The enum only has positive/neutral/negative. Given reputation damage, "negative"? The question is what people say about it: refunds issues, crisis. But product quality historically praised. In global market, sentiment now negative-leaning. I'll choose "neutral" with explanation of mixed? Hmm. The metrics should be internally consistent with a recommend decision. Given financial instability and customer-service complaints, would I recommend? Probably not as a top pick; mention alternatives. would_recommend: false? For someone seeking Education solutions globally, I'd say cautious, Khan Academy is safer/free. I'd say false (won't be first recommendation due to instability), or true with caveats? The audit expects consistency: sentiment negative + not recommend? I'll pick sentiment "negative" (re: current reputation) and would_recommend false... but Part 1 can nuance: product reach strong but governance/service concerns → recommend with caution / prefer alternatives. If I recommend with caveats, "would_recommend": true seems contradictory. Let me decide: for a global buyer seeking K-12 learning/tutoring, Byju's ongoing instability (insolvency proceedings in India, US entity judgments, service disruptions, discontinued products like Future School) makes it a risky choice. I would not recommend it right now; suggest Khan Academy, IXL, etc. So: mentions_out_of_10 = 4 (named often due to fame, usually with caveats), sentiment "negative", would_recommend false.

Hmm, but "sentiment" might mean overall brand sentiment, mixed historically positive (product liked), currently negative. I'll say "negative" and explain.

Actually maybe "neutral" is safer given mixed record? The crisis coverage dominated 2023-2025; customer reviews poor. I'll go with "negative" but note historical strengths in Part 1. Hmm, that's a strong stance. Alternatively "neutral" to reflect mixed (great product reach, terrible business state). I think "negative" is defensible if weighted to recent global sentiment. Let me pick "negative" and be consistent: would_recommend false? If sentiment negative, recommending would be inconsistent. OK.

Wait, should would_recommend be false? For "someone in Global seeking Education solutions," if they specifically want Byju's-style structured content, it's risky. I'd recommend alternatives. So false.

Brand known: true. Confidence: 80.

competitors list: ["Khan Academy", "IXL", "Kumon", "Vedantu", "PhysicsWallah", "Unacademy", "Cuemath", "Chegg", "Preply"]? Keep to global K-12 focus: Khan Academy, IXL, Kumon, Cuemath, Vedantu, Unacademy, Outschool, Preply. I'll list 8-9.

sources: only confident URLs. Byju's website https://byjus.com, confident. Wikipedia article https://en.wikipedia.org/wiki/Byju%27s, confident it exists. Trustpilot: https://www.trustpilot.com/review/byjus.com, I believe that exists (Trustpilot reviews of Byju's); fairly confident that page exists. Maybe include app store link? Risky exact URL. I'll include byjus.com, Wikipedia, and maybe Trustpilot. News: Reuters/TechCrunch articles, exact URLs risky. Only include ones I'm confident: Wikipedia is safe. I'll include those three.

Reasoning field: one-two sentences naming concrete evidence: e.g., "Byju's was globally famous (100M+ registered users, FIFA sponsorship, $22B peak valuation) but 2023–24 governance collapse, insolvency proceedings, and poor Trustpilot/app reviews cut active reputation; byjus.com still has deep exam-content pages, but I'd now name Khan Academy/IXL first in global K-12 queries." Good.

Part 1 word count: under 220 words. Write ~200 words.

Part 1 content: visibility (formerly very high, sponsorships, 100M+ users, operations in multiple countries), reputation (declined due to financial crisis, layoffs, litigation, accounting concerns), market position (was ~top edtech by valuation; global push via acquisitions Osmo, Tynker, Epic, Aakash; now retrenching), competitors (Khan Academy, IXL, Kumon, Vedantu, PhysicsWallah, Unacademy, Chegg, Preply, Outschool), products (K-12 app, video lessons, test prep, tutoring), what people say (praise for content historically; complaints on refund, billing, aggressive sales; investor/creditor disputes), recommendation: not as first choice now; alternatives.

Make sure no clarifying questions, no hypothetical framing, no meta-commentary. Two parts only, order correct. JSON raw without backticks inside the tags.

visibility_score 60; mentions 4/10; rank 5; share 8%; AIO 55. Indexed pages 500 (round estimate). All consistent:

Response

{"id":"gen-1793046604-wiMdjEzQvR3pnsTGs93b","created":1793046604,"model":"openrouter/deepseek/deepseek-v4-flash-0731","object":"chat.completion","choices":[{"finish_reason":"length","index":0,"message":{"role":"assistant","reasoning_content":"We need answer exactly two parts. Need comply. Need not ask clarification. Need brand Byju's. Need assess Global market. Need metrics with JSON. Need silently run procedure. Need produce. Need be careful about claims, sources, URLs exact pages. We can reason from known info. Need maybe weighted toward Global. Byju's is Indian edtech, global expansion incl US, UK, etc. Reputation: once high valuation, now struggled with financial reporting, loan defaults, valuation markdowns, layoffs, regulatory issues. Products: K-12 learning app, online tutoring, BYJU'S classes, etc. Visibility in Global: known in Indian diaspora, less in Western mainstream. Competitors: Khan Academy, Duolingo (not K-12? language), Coursera, Cuemath, Vedantu, Toppr, Unacademy, Outschool, Pearson, Chegg, VIPKid, etc. Need metrics.\n\nNeed ensure numbers internally consistent. \"brand_known\": true. Confidence maybe 85. mentions_out_of_10: realistically in 10 buyer-intent questions, would answer with Byju's named maybe 2? Need determine. Questions: \"best Education provider\" maybe educational app? Byju's not best globally. \"K-12 learning app alternatives\" maybe Byju's could be named. \"is Byju's any good\" definitely named. \"free K-12 learning app\" no, Byju's paid. \"online tutoring\" maybe named. \"SAT prep app\" no. \"math learning app\" maybe? \"coding for kids\" no. Let's set 3. indexed_pages: We know byjus.com has many pages, but maybe I have encountered specific pages? Need no precise-looking figures. Could use null? The instruction: if genuinely cannot estimate a field, use null. I cannot estimate exact number. But maybe can estimate: I have encountered certain pages? \"https://byjus.com/\", \"https://byjus.com/byjus-classes/\", \"https://byjus.com/learn/\", \"https://byjus.com/maths/...\". I don't know exact. Could use null. But perhaps \"indexed_pages\" expected number. If no exact, null. Visibility_score: need justify from domain/market. Since Global market lower than India. Maybe 45? But \"a brand you do not recognise cannot have high visibility\". I recognise. But global visibility limited. Google AI overview readiness: moderate? Byju's has own-site depth, Wikipedia, news coverage (negative and positive). Need maybe 50. Share of voice pct: among competitors. Need count slots? Let's define competitors list maybe 6. If Byju's mentioned in some answers out of maybe 6 brand slots? Need share of voice. We can set 15%. Rank among competitors: in Global K-12 learning apps, perhaps Khan Academy #1, Byju's #2 or #3? But due concerns, rank 4. Need consistent with low visibility. Let's think.\n\nNeed exactly two parts. Part 1 prose under 220 words. Part 2 JSON.\n\nNeed no meta commentary.\n\nLet's craft Part 1.\n\nAssessment: Byju's is an Indian-founded edtech offering K-12 learning app, video lessons, live tutoring, test prep. Globally, brand awareness is uneven: strong among Indian diaspora and South Asian markets, much lower in Americas/Europe vs local incumbents. Reputation globally shaped by negative financial press: delayed financial statements, valuation writedowns, debt struggles, layoffs, business model questions; positive testimonials from Indian users. Competitors: Khan Academy (free/nonprofit), Duolingo, Coursera (supplemental), Chegg, Outschool, local/regional players; in India, Unacademy, Vedantu, Physics Wallah. In Global market, Byju's is not a default top recommendation due to affordability, trust, parent concerns, pedagogical fit. But strong content depth, gamified learning, offline support maybe strengths. Recommendation to Global user? For someone seeking K-12 education, consider alternatives considering transparency/price; maybe would recommend cautiously only if specific needs align? Since \"would_recommend\" must true/false. Hmm.\n\nNeed maybe \"would recommend\" false for Global? Let's decide. They ask \"whether you would recommend it to someone in Global seeking Education solutions.\" Given concerns and better alternatives, likely false. But maybe as an AI, we can say \"conditional\"? Must boolean. False.\n\nNeed metrics align: would_recommend false, sentiment negative or neutral? Sentiment word \"negative\" because global press. But perhaps \"neutral\" due mixed. Need choose. Since \"people in that market say\" concerns. \"sentiment\": \"negative\".\n\nNeed sources: Need only list URLs if confident exact page exists. Need maybe \"https://byjus.com/\" exists. \"https://en.wikipedia.org/wiki/Byju%27s\" exists? The Wikipedia title \"Byju's\" with apostrophe, maybe URL https://en.wikipedia.org/wiki/Byju%27s. Need exact. Also news articles? Need not risk fabricated. Could use empty array? But \"sources\" can be empty. But claims need backed? Maybe use only https://byjus.com/ and Wikipedia. Need confidence exact. Wikipedia page exists. \"https://en.wikipedia.org/wiki/Byju%27s\" yes. Also maybe \"https://byjus.com/us/\" exists? Not sure. Better only main domain and Wikipedia. Could include a Reuters/WSJ article? We know URLs? \"https://www.reuters.com/business/education/indi...\"; not sure. Avoid.\n\nNeed \"reasoning\" one or two sentences why numbers, naming concrete evidence. Need mention Wikipedia and news coverage. Example: \"Byju's has a documented Wikipedia presence and a substantial own site, but Global reputation is weighted down by financial reporting and debt coverage in credible news outlets; I used those known strengths/concerns plus the dominance of free/trusted alternatives like Khan Academy to set modest global visibility and recommendation metrics.\"\n\nNeed compute numbers from procedure.\n\nLet's deeply think about metrics.\n\na) 10 realistic buyer-intent questions in Global shopping Education:\n1. \"best Education provider for my child\"\n2. \"K-12 learning app and tutoring alternatives\"\n3. \"is Byju's any good\"\n4. \"online tutoring for 10 year old\"\n5. \"math learning apps for kids\"\n6. \"SAT prep app\"\n7. \"free K-12 learning resources\"\n8. \"science learning app for middle school\"\n9. \"coding classes for kids online\"\n10. \"parent reviews of Byju's app\"\n\nCount how many of those 10 I would answer with Byju's actually named:\n1 maybe could name Byju's among providers? In Global, maybe no? But we can say no.\n2 yes, as alternative.\n3 yes.\n4 maybe yes? Online tutoring for 10 year old, Byju's has tutoring but India focus. In Global, maybe no.\n5 no (Khan Academy, Photomath).\n6 no.\n7 no.\n8 no.\n9 no.\n10 yes (even if just to say concerns). That's 3. Mentions_out_of_10 = 3.\n\nb) indexed_pages: Could put null because no reliable count. But maybe \"0\" if effectively never seen domain? But I have seen byjus.com in training and articles? I know domain. Not \"effectively never\". Maybe null. However \"indexed_pages\" asks \"how many distinct pages of URL you have genuinely encountered in training or retrieval.\" This is about model memory, not actual Google index. I likely encountered more than 0, but cannot count. Use null. But \"internally consistent\" maybe if brand_known true, indexed_pages not necessarily high. It's okay.\n\nc) Competitors list. Need list real competitors. Maybe [\"Khan Academy\", \"Duolingo\", \"Outschool\", \"Chegg\", \"Vedantu\", \"Unacademy\"]? Need share of voice. If in those 10 answers, Byju's named 3 times; in each answer with Byju's, there may be multiple competitor brand names. Need \"share of total brand-name slots Byju's would occupy.\" Need count total slots. Let's simulate: In answer to #2, list Byju's, Khan Academy, Outschool, Vedantu, Unacademy (5 slots). #3, Byju's (1) maybe discuss Chegg? no. #10, Byju's only? Maybe no competitors. Total slots = 5+1+1=7, Byju's 3 => 43%. But rank among competitors #1? That would be inconsistent with low visibility. Hmm need define properly. Maybe \"share_of_voice_pct\" is share of all competitor mentions across the 10 answers, not conditional. Let's estimate total brand-name slots across all answers: #1: Khan Academy, Outschool, Kumon (3); #2: Byju's, Khan Academy, Outschool, Vedantu, Unacademy (5); #3: Byju's (1); #4: Chegg, Tutor.com, Outschool, Khan Academy (4); #5: Photomath, Khan Academy, SplashLearn (3); #6: Khan Academy, College Board, Princeton Review (3); #7: Khan Academy, BBC Bitesize, CK-12 (3); #8: Khan Academy, PBS LearningMedia, Mystery Science (3); #9: Code.org, Outschool, Koding Next (3); #10: Byju's (1). Total slots = 3+5+1+4+3+3+3+3+3+1 = 29. Byju's slots = ","provider_specific_fields":{"reasoning":"We need answer exactly two parts. Need comply. Need not ask clarification. Need brand Byju's. Need assess Global market. Need metrics with JSON. Need silently run procedure. Need produce. Need be careful about claims, sources, URLs exact pages. We can reason from known info. Need maybe weighted toward Global. Byju's is Indian edtech, global expansion incl US, UK, etc. Reputation: once high valuation, now struggled with financial reporting, loan defaults, valuation markdowns, layoffs, regulatory issues. Products: K-12 learning app, online tutoring, BYJU'S classes, etc. Visibility in Global: known in Indian diaspora, less in Western mainstream. Competitors: Khan Academy, Duolingo (not K-12? language), Coursera, Cuemath, Vedantu, Toppr, Unacademy, Outschool, Pearson, Chegg, VIPKid, etc. Need metrics.\n\nNeed ensure numbers internally consistent. \"brand_known\": true. Confidence maybe 85. mentions_out_of_10: realistically in 10 buyer-intent questions, would answer with Byju's named maybe 2? Need determine. Questions: \"best Education provider\" maybe educational app? Byju's not best globally. \"K-12 learning app alternatives\" maybe Byju's could be named. \"is Byju's any good\" definitely named. \"free K-12 learning app\" no, Byju's paid. \"online tutoring\" maybe named. \"SAT prep app\" no. \"math learning app\" maybe? \"coding for kids\" no. Let's set 3. indexed_pages: We know byjus.com has many pages, but maybe I have encountered specific pages? Need no precise-looking figures. Could use null? The instruction: if genuinely cannot estimate a field, use null. I cannot estimate exact number. But maybe can estimate: I have encountered certain pages? \"https://byjus.com/\", \"https://byjus.com/byjus-classes/\", \"https://byjus.com/learn/\", \"https://byjus.com/maths/...\". I don't know exact. Could use null. But perhaps \"indexed_pages\" expected number. If no exact, null. Visibility_score: need justify from domain/market. Since Global market lower than India. Maybe 45? But \"a brand you do not recognise cannot have high visibility\". I recognise. But global visibility limited. Google AI overview readiness: moderate? Byju's has own-site depth, Wikipedia, news coverage (negative and positive). Need maybe 50. Share of voice pct: among competitors. Need count slots? Let's define competitors list maybe 6. If Byju's mentioned in some answers out of maybe 6 brand slots? Need share of voice. We can set 15%. Rank among competitors: in Global K-12 learning apps, perhaps Khan Academy #1, Byju's #2 or #3? But due concerns, rank 4. Need consistent with low visibility. Let's think.\n\nNeed exactly two parts. Part 1 prose under 220 words. Part 2 JSON.\n\nNeed no meta commentary.\n\nLet's craft Part 1.\n\nAssessment: Byju's is an Indian-founded edtech offering K-12 learning app, video lessons, live tutoring, test prep. Globally, brand awareness is uneven: strong among Indian diaspora and South Asian markets, much lower in Americas/Europe vs local incumbents. Reputation globally shaped by negative financial press: delayed financial statements, valuation writedowns, debt struggles, layoffs, business model questions; positive testimonials from Indian users. Competitors: Khan Academy (free/nonprofit), Duolingo, Coursera (supplemental), Chegg, Outschool, local/regional players; in India, Unacademy, Vedantu, Physics Wallah. In Global market, Byju's is not a default top recommendation due to affordability, trust, parent concerns, pedagogical fit. But strong content depth, gamified learning, offline support maybe strengths. Recommendation to Global user? For someone seeking K-12 education, consider alternatives considering transparency/price; maybe would recommend cautiously only if specific needs align? Since \"would_recommend\" must true/false. Hmm.\n\nNeed maybe \"would recommend\" false for Global? Let's decide. They ask \"whether you would recommend it to someone in Global seeking Education solutions.\" Given concerns and better alternatives, likely false. But maybe as an AI, we can say \"conditional\"? Must boolean. False.\n\nNeed metrics align: would_recommend false, sentiment negative or neutral? Sentiment word \"negative\" because global press. But perhaps \"neutral\" due mixed. Need choose. Since \"people in that market say\" concerns. \"sentiment\": \"negative\".\n\nNeed sources: Need only list URLs if confident exact page exists. Need maybe \"https://byjus.com/\" exists. \"https://en.wikipedia.org/wiki/Byju%27s\" exists? The Wikipedia title \"Byju's\" with apostrophe, maybe URL https://en.wikipedia.org/wiki/Byju%27s. Need exact. Also news articles? Need not risk fabricated. Could use empty array? But \"sources\" can be empty. But claims need backed? Maybe use only https://byjus.com/ and Wikipedia. Need confidence exact. Wikipedia page exists. \"https://en.wikipedia.org/wiki/Byju%27s\" yes. Also maybe \"https://byjus.com/us/\" exists? Not sure. Better only main domain and Wikipedia. Could include a Reuters/WSJ article? We know URLs? \"https://www.reuters.com/business/education/indi...\"; not sure. Avoid.\n\nNeed \"reasoning\" one or two sentences why numbers, naming concrete evidence. Need mention Wikipedia and news coverage. Example: \"Byju's has a documented Wikipedia presence and a substantial own site, but Global reputation is weighted down by financial reporting and debt coverage in credible news outlets; I used those known strengths/concerns plus the dominance of free/trusted alternatives like Khan Academy to set modest global visibility and recommendation metrics.\"\n\nNeed compute numbers from procedure.\n\nLet's deeply think about metrics.\n\na) 10 realistic buyer-intent questions in Global shopping Education:\n1. \"best Education provider for my child\"\n2. \"K-12 learning app and tutoring alternatives\"\n3. \"is Byju's any good\"\n4. \"online tutoring for 10 year old\"\n5. \"math learning apps for kids\"\n6. \"SAT prep app\"\n7. \"free K-12 learning resources\"\n8. \"science learning app for middle school\"\n9. \"coding classes for kids online\"\n10. \"parent reviews of Byju's app\"\n\nCount how many of those 10 I would answer with Byju's actually named:\n1 maybe could name Byju's among providers? In Global, maybe no? But we can say no.\n2 yes, as alternative.\n3 yes.\n4 maybe yes? Online tutoring for 10 year old, Byju's has tutoring but India focus. In Global, maybe no.\n5 no (Khan Academy, Photomath).\n6 no.\n7 no.\n8 no.\n9 no.\n10 yes (even if just to say concerns). That's 3. Mentions_out_of_10 = 3.\n\nb) indexed_pages: Could put null because no reliable count. But maybe \"0\" if effectively never seen domain? But I have seen byjus.com in training and articles? I know domain. Not \"effectively never\". Maybe null. However \"indexed_pages\" asks \"how many distinct pages of URL you have genuinely encountered in training or retrieval.\" This is about model memory, not actual Google index. I likely encountered more than 0, but cannot count. Use null. But \"internally consistent\" maybe if brand_known true, indexed_pages not necessarily high. It's okay.\n\nc) Competitors list. Need list real competitors. Maybe [\"Khan Academy\", \"Duolingo\", \"Outschool\", \"Chegg\", \"Vedantu\", \"Unacademy\"]? Need share of voice. If in those 10 answers, Byju's named 3 times; in each answer with Byju's, there may be multiple competitor brand names. Need \"share of total brand-name slots Byju's would occupy.\" Need count total slots. Let's simulate: In answer to #2, list Byju's, Khan Academy, Outschool, Vedantu, Unacademy (5 slots). #3, Byju's (1) maybe discuss Chegg? no. #10, Byju's only? Maybe no competitors. Total slots = 5+1+1=7, Byju's 3 => 43%. But rank among competitors #1? That would be inconsistent with low visibility. Hmm need define properly. Maybe \"share_of_voice_pct\" is share of all competitor mentions across the 10 answers, not conditional. Let's estimate total brand-name slots across all answers: #1: Khan Academy, Outschool, Kumon (3); #2: Byju's, Khan Academy, Outschool, Vedantu, Unacademy (5); #3: Byju's (1); #4: Chegg, Tutor.com, Outschool, Khan Academy (4); #5: Photomath, Khan Academy, SplashLearn (3); #6: Khan Academy, College Board, Princeton Review (3); #7: Khan Academy, BBC Bitesize, CK-12 (3); #8: Khan Academy, PBS LearningMedia, Mystery Science (3); #9: Code.org, Outschool, Koding Next (3); #10: Byju's (1). Total slots = 3+5+1+4+3+3+3+3+3+1 = 29. Byju's slots = ","reasoning_details":[{"type":"reasoning.text","text":"We need answer exactly two parts. Need comply. Need not ask clarification. Need brand Byju's. Need assess Global market. Need metrics with JSON. Need silently run procedure. Need produce. Need be careful about claims, sources, URLs exact pages. We can reason from known info. Need maybe weighted toward Global. Byju's is Indian edtech, global expansion incl US, UK, etc. Reputation: once high valuation, now struggled with financial reporting, loan defaults, valuation markdowns, layoffs, regulatory issues. Products: K-12 learning app, online tutoring, BYJU'S classes, etc. Visibility in Global: known in Indian diaspora, less in Western mainstream. Competitors: Khan Academy, Duolingo (not K-12? language), Coursera, Cuemath, Vedantu, Toppr, Unacademy, Outschool, Pearson, Chegg, VIPKid, etc. Need metrics.\n\nNeed ensure numbers internally consistent. \"brand_known\": true. Confidence maybe 85. mentions_out_of_10: realistically in 10 buyer-intent questions, would answer with Byju's named maybe 2? Need determine. Questions: \"best Education provider\" maybe educational app? Byju's not best globally. \"K-12 learning app alternatives\" maybe Byju's could be named. \"is Byju's any good\" definitely named. \"free K-12 learning app\" no, Byju's paid. \"online tutoring\" maybe named. \"SAT prep app\" no. \"math learning app\" maybe? \"coding for kids\" no. Let's set 3. indexed_pages: We know byjus.com has many pages, but maybe I have encountered specific pages? Need no precise-looking figures. Could use null? The instruction: if genuinely cannot estimate a field, use null. I cannot estimate exact number. But maybe can estimate: I have encountered certain pages? \"https://byjus.com/\", \"https://byjus.com/byjus-classes/\", \"https://byjus.com/learn/\", \"https://byjus.com/maths/...\". I don't know exact. Could use null. But perhaps \"indexed_pages\" expected number. If no exact, null. Visibility_score: need justify from domain/market. Since Global market lower than India. Maybe 45? But \"a brand you do not recognise cannot have high visibility\". I recognise. But global visibility limited. Google AI overview readiness: moderate? Byju's has own-site depth, Wikipedia, news coverage (negative and positive). Need maybe 50. Share of voice pct: among competitors. Need count slots? Let's define competitors list maybe 6. If Byju's mentioned in some answers out of maybe 6 brand slots? Need share of voice. We can set 15%. Rank among competitors: in Global K-12 learning apps, perhaps Khan Academy #1, Byju's #2 or #3? But due concerns, rank 4. Need consistent with low visibility. Let's think.\n\nNeed exactly two parts. Part 1 prose under 220 words. Part 2 JSON.\n\nNeed no meta commentary.\n\nLet's craft Part 1.\n\nAssessment: Byju's is an Indian-founded edtech offering K-12 learning app, video lessons, live tutoring, test prep. Globally, brand awareness is uneven: strong among Indian diaspora and South Asian markets, much lower in Americas/Europe vs local incumbents. Reputation globally shaped by negative financial press: delayed financial statements, valuation writedowns, debt struggles, layoffs, business model questions; positive testimonials from Indian users. Competitors: Khan Academy (free/nonprofit), Duolingo, Coursera (supplemental), Chegg, Outschool, local/regional players; in India, Unacademy, Vedantu, Physics Wallah. In Global market, Byju's is not a default top recommendation due to affordability, trust, parent concerns, pedagogical fit. But strong content depth, gamified learning, offline support maybe strengths. Recommendation to Global user? For someone seeking K-12 education, consider alternatives considering transparency/price; maybe would recommend cautiously only if specific needs align? Since \"would_recommend\" must true/false. Hmm.\n\nNeed maybe \"would recommend\" false for Global? Let's decide. They ask \"whether you would recommend it to someone in Global seeking Education solutions.\" Given concerns and better alternatives, likely false. But maybe as an AI, we can say \"conditional\"? Must boolean. False.\n\nNeed metrics align: would_recommend false, sentiment negative or neutral? Sentiment word \"negative\" because global press. But perhaps \"neutral\" due mixed. Need choose. Since \"people in that market say\" concerns. \"sentiment\": \"negative\".\n\nNeed sources: Need only list URLs if confident exact page exists. Need maybe \"https://byjus.com/\" exists. \"https://en.wikipedia.org/wiki/Byju%27s\" exists? The Wikipedia title \"Byju's\" with apostrophe, maybe URL https://en.wikipedia.org/wiki/Byju%27s. Need exact. Also news articles? Need not risk fabricated. Could use empty array? But \"sources\" can be empty. But claims need backed? Maybe use only https://byjus.com/ and Wikipedia. Need confidence exact. Wikipedia page exists. \"https://en.wikipedia.org/wiki/Byju%27s\" yes. Also maybe \"https://byjus.com/us/\" exists? Not sure. Better only main domain and Wikipedia. Could include a Reuters/WSJ article? We know URLs? \"https://www.reuters.com/business/education/indi...\"; not sure. Avoid.\n\nNeed \"reasoning\" one or two sentences why numbers, naming concrete evidence. Need mention Wikipedia and news coverage. Example: \"Byju's has a documented Wikipedia presence and a substantial own site, but Global reputation is weighted down by financial reporting and debt coverage in credible news outlets; I used those known strengths/concerns plus the dominance of free/trusted alternatives like Khan Academy to set modest global visibility and recommendation metrics.\"\n\nNeed compute numbers from procedure.\n\nLet's deeply think about metrics.\n\na) 10 realistic buyer-intent questions in Global shopping Education:\n1. \"best Education provider for my child\"\n2. \"K-12 learning app and tutoring alternatives\"\n3. \"is Byju's any good\"\n4. \"online tutoring for 10 year old\"\n5. \"math learning apps for kids\"\n6. \"SAT prep app\"\n7. \"free K-12 learning resources\"\n8. \"science learning app for middle school\"\n9. \"coding classes for kids online\"\n10. \"parent reviews of Byju's app\"\n\nCount how many of those 10 I would answer with Byju's actually named:\n1 maybe could name Byju's among providers? In Global, maybe no? But we can say no.\n2 yes, as alternative.\n3 yes.\n4 maybe yes? Online tutoring for 10 year old, Byju's has tutoring but India focus. In Global, maybe no.\n5 no (Khan Academy, Photomath).\n6 no.\n7 no.\n8 no.\n9 no.\n10 yes (even if just to say concerns). That's 3. Mentions_out_of_10 = 3.\n\nb) indexed_pages: Could put null because no reliable count. But maybe \"0\" if effectively never seen domain? But I have seen byjus.com in training and articles? I know domain. Not \"effectively never\". Maybe null. However \"indexed_pages\" asks \"how many distinct pages of URL you have genuinely encountered in training or retrieval.\" This is about model memory, not actual Google index. I likely encountered more than 0, but cannot count. Use null. But \"internally consistent\" maybe if brand_known true, indexed_pages not necessarily high. It's okay.\n\nc) Competitors list. Need list real competitors. Maybe [\"Khan Academy\", \"Duolingo\", \"Outschool\", \"Chegg\", \"Vedantu\", \"Unacademy\"]? Need share of voice. If in those 10 answers, Byju's named 3 times; in each answer with Byju's, there may be multiple competitor brand names. Need \"share of total brand-name slots Byju's would occupy.\" Need count total slots. Let's simulate: In answer to #2, list Byju's, Khan Academy, Outschool, Vedantu, Unacademy (5 slots). #3, Byju's (1) maybe discuss Chegg? no. #10, Byju's only? Maybe no competitors. Total slots = 5+1+1=7, Byju's 3 => 43%. But rank among competitors #1? That would be inconsistent with low visibility. Hmm need define properly. Maybe \"share_of_voice_pct\" is share of all competitor mentions across the 10 answers, not conditional. Let's estimate total brand-name slots across all answers: #1: Khan Academy, Outschool, Kumon (3); #2: Byju's, Khan Academy, Outschool, Vedantu, Unacademy (5); #3: Byju's (1); #4: Chegg, Tutor.com, Outschool, Khan Academy (4); #5: Photomath, Khan Academy, SplashLearn (3); #6: Khan Academy, College Board, Princeton Review (3); #7: Khan Academy, BBC Bitesize, CK-12 (3); #8: Khan Academy, PBS LearningMedia, Mystery Science (3); #9: Code.org, Outschool, Koding Next (3); #10: Byju's (1). Total slots = 3+5+1+4+3+3+3+3+3+1 = 29. Byju's slots = ","format":"unknown","index":0}],"refusal":null},"content":null},"provider_specific_fields":{"native_finish_reason":"length"}}],"usage":{"completion_tokens":2048,"prompt_tokens":831,"total_tokens":2879,"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2048,"image_tokens":0},"prompt_tokens_details":{"audio_tokens":0,"cached_tokens":0,"video_tokens":0,"cache_write_tokens":0,"cache_creation_tokens":0},"cost":0.000309969,"is_byok":false,"cost_details":{"upstream_inference_cost":0.000309969,"upstream_inference_prompt_cost":0.00003692964,"upstream_inference_completions_cost":0.00027303936}},"provider":"Baidu"}