BookSync Integration Research

Author: Kai, 2026-04-18 Context: David asked "how do we add a book sync, do we need API keys, do we need to contact the books, how much does it cost, draft so we're ready." Decision needed: which BookSync path does NuroPicks commit to, and when?

TL;DR

Sportsbooks do NOT offer consumer-account APIs. Pikkit's BookSync is browser-automation with locally-stored credentials, NOT an API integration. We have four real paths, ranked best-first for our stage:

My recommendation: ship #1 at launch, test #2 in parallel, plan #3 for month 6+.

1. Sportsbook API landscape (what's actually available)

Odds APIs (GOOD, used for picks)

• The Odds API ($59/mo) — live odds across 150+ books. What we're already planning for picks.
• OddsJam API (enterprise pricing) — same category.
• SharpRank API — includes odds + some advanced features.

None of these give access to a user's personal bet history at their own sportsbook. They're for market data.

Consumer account APIs (DOES NOT EXIST at retail US books)

Hit the developer portals for the major US sportsbooks 2026-04-18:

• DraftKings: No public consumer-account API. They have an "Affiliates API" for referral tracking but not bet history.
• FanDuel / Flutter: No public API. Internal APIs are protected.
• BetMGM / Entain: No public API. Partnership track requires multi-million-dollar revenue share with MGM Resorts.
• Caesars: Same. Affiliate programs only.
• Fanatics Sportsbook: No API.
• BetRivers / SugarHouse: No API.
• Circa: No API (sharp book, fewer partners).
• Pinnacle: No public consumer API (offshore).

Conclusion: asking sportsbooks for API access is a dead end. They don't offer it, and even if they did, the cost would be prohibitive (Pikkit doesn't use official APIs for a reason — there aren't any).

2. How Pikkit actually does it

From their public FAQ (scraped 2026-04-18):

"Pikkit works by using BookSync to easily log in to your favorite sportsbooks and sync your bets instantly."

"Account credentials are stored locally with the help of system keychain (Apple Keychain/Credential Vault), similar to how a browser stores your passwords."

Translation: Pikkit asks users for their sportsbook email + password, stores them in the phone's secure keystore (NOT on Pikkit's servers), and uses those credentials to log in via WebView or session-based scraping to pull bet history. This is how Mint/Plaid/Yodlee work for banking, except:

1. Banks have the EFTA/NACHA regulatory framework that allows for aggregator access
2. Sportsbooks have NO such framework; each one's ToS typically says "you may not share credentials or use automated tools"
3. When a sportsbook detects automated scraping of account data, they can void the account + keep funds

Pikkit gets away with this because:

• They're an iOS-first app (session lives on-device)
• Scale is modest (hundreds of thousands of users, not millions)
• Books haven't aggressively enforced because Pikkit drives affiliate traffic to them
• $75M+ in VC funding behind them (Matrix Partners, etc.) buys legal team

3. Our four implementation paths

Path 1: Manual /bet + OCR screenshot import

What: User either types their bet into our /bet Discord command, or uploads a screenshot of the betslip which we OCR via Gemini Vision (Nano Banana + text extraction) and pre-fill the /bet form.

Cost:

• Engineering: 1-2 weeks to wire OCR into /bet + web
• Runtime: Gemini Vision is pay-per-call, ~$0.01 per screenshot, so at 1000 daily imports = $10/day = $300/mo
• At Pro+Elite subscribers only: likely $50-150/mo real runtime cost

Timeline: days to MVP, 2 weeks to polish.

Legal: zero risk. Users are voluntarily uploading their own data.

Pro:

• Ships immediately
• No credential storage means no account-takeover attack surface
• We own the data quality (our parser, our normalization)
• Users teach the OCR model over time via corrections

Con:

• Worse UX than Pikkit's auto-sync
• Users won't backfill huge historical samples
• Power users will churn to Pikkit for convenience

Path 2: SharpRank BetSync license

What: SharpRank (the odds/analytics API vendor) has a product called BetSync that does credential-based bet-history sync. They license it to developers.

Cost: unknown publicly. Estimate based on similar consumer-banking aggregators:

• Per-connected-book fee: $0.50-2/user/mo
• Or rev share: 5-15% of user subscription
• Plus a yearly platform fee ($5k-25k) to gain access
• At 500 paying users across 3 connected books: ~$750-3000/mo direct cost

Timeline: 1-2 weeks if they'll partner with us. Risk: they may refuse to license to a direct Pikkit competitor.

Legal: low risk for us. SharpRank carries the compliance burden. We just pay them.

Pro:

• Fastest path to Pikkit-parity on BookSync
• They handle credential security + book-side ToS issues
• Book coverage probably 15-25 of the top 30

Con:

• Recurring margin cost per user
• They could pull access at any time
• Not a moat; any competitor can license the same thing

Path 3: Build our own session-based sync (Pikkit clone)

What: We build native iOS + Android apps (or headless browser automation) that log into sportsbooks on the user's behalf and scrape bet history. Credentials stored in device secure keystore only.

Cost:

• Engineering: 2-3 months for a senior mobile + backend engineer ($15-30k fully loaded)
• Legal opinion on ToS + CFAA (Computer Fraud and Abuse Act) compliance: $5-15k one-time
• Ongoing maintenance: scrapers break when books update their login flow. Budget $2-4k/mo in eng time (~20-40 hrs/mo).
• App Store compliance review: 2-4 weeks for first submission

Timeline: 3-6 months end-to-end.

Legal risk: medium-high.

• CFAA compliance: the legal question is whether scraping with user credentials counts as "unauthorized access." The hiQ Labs vs LinkedIn case (2021) narrowed CFAA to explicit bypass-of-technical-barriers, which we wouldn't do. Users voluntarily share their own credentials with us.
• Sportsbook ToS violation: most sportsbook ToS forbid "sharing credentials" or "automated access." Our counter: the user is authorizing access and we're not manipulating odds/bets, only reading history. Still a risk the book could void a user account.
• Multi-state gambling compliance: probably not a compliance hit since we're not accepting wagers, just reading bet history.

Pro:

• Closest to Pikkit's moat; a real differentiator
• Zero recurring license fees (we own the stack)
• Ability to sync ANY book, not just SharpRank's roster
• Marketing: "NuroPicks is the only platform with auto-sync + AI picks + capper marketplace"

Con:

• Large upfront eng cost
• Ongoing maintenance tax (estimated 15-25% of a single eng's time forever)
• Credential storage is a huge attack surface — one breach wipes trust
• Each book update can break sync for days to weeks; customer support load spikes

Path 4: Deep-link bet CONFIRMATION scheme

What: User starts bet on our side (parlay builder), we generate a deep link to the sportsbook, user places the bet. Some books (DraftKings, FanDuel) can return the bet confirmation via a callback URL or event bus.

Cost: low eng (1 week per book integration, 4-6 books viable).

Timeline: 1-2 months.

Legal: clean. Users voluntarily confirm bets to us, no credential storage.

Pro:

• Legitimate book-approved path (some books welcome this)
• No credential security burden
• Pairs naturally with our parlay builder

Con:

• Only captures bets the user started on our platform, not historical sync
• Book coverage limited to those with callback schemes (DK + FD + BetMGM maybe)
• Doesn't solve the "import my full Pikkit history" use case

4. Recommended sequencing

Launch (week 0-4): Path 1 + Path 4

• Manual /bet entry (already built) + screenshot OCR import
• Parlay builder with deep-link confirmation for DK/FD (captures bet placement)
• Marketing line: "we respect your sportsbook accounts, we don't ask for your passwords"
• Pros: ships now, zero legal risk, differentiates us from Pikkit's security-surface story

Month 2-3: Test Path 2

• Reach out to SharpRank about BetSync license terms
• Test integration with 100 beta users
• Decide at month 3 whether economics work (per-user margin + refund rate)

Month 6+: Path 3 if we have scale + capital

• Once we have 500+ paying users and can justify the ~3 month engineering investment
• Get the legal opinion upfront before any user credential touches our system
• Beta with 20 users in an isolated CI environment before production
• Market it as "NuroPicks Sync" with strong security story

5. Client contact checklist

When we're ready to reach out to partners:

SharpRank (Path 2)

• Email: partnerships@sharprank.com
• LinkedIn: [search for their BD/partnerships lead]
• Angle: "we're a sports-betting-intelligence platform looking to integrate bet sync"
• Ask: pricing, book coverage, exclusivity terms, revenue share model

Direct sportsbooks (Affiliate programs, not API)

• DraftKings: dkpartners@draftkings.com (affiliate/marketing only)
• FanDuel: partners@fanduel.com
• BetMGM: partnerships@betmgm.com
• Ask: affiliate revenue share + deep-link support + co-marketing opportunities. Do NOT ask about consumer account APIs, they don't exist.

Pikkit (unlikely to license but worth trying)

• support@pikkit.app
• Pitch: sub-license BookSync under NDA for a non-directly-competitive vertical (we focus on AI picks, they focus on social tracking)
• Likely response: no, but worth a 30-min call

6. Cost summary (David's budget question)

Launch budget (Paths 1 + 4):

• Gemini Vision OCR: ~$50-150/mo runtime
• Deep-link scheme integration: one-time eng, no recurring
• Total: ~$50-150/mo at launch scale

Path 2 budget if we license SharpRank:

• ~$0.50-2/user/mo * connected users
• At 500 Elite subs * 2 books average = 1000 connections * $1 = $1000/mo
• Passed through in the $79.99 Elite price tier

Path 3 budget if we build our own:

• $15-30k upfront engineering
• $5-15k legal opinion
• $2-4k/mo ongoing maintenance
• Break-even: 500+ Elite subs to justify vs paying SharpRank

7. Decision point for David

Short-term (this week): approve Path 1 launch scope.

• OCR screenshot import for /bet
• Manual /bet entry UX polish
• Deep-link parlay builder (already in our scope post-odds-key)

Medium-term (month 2): research Path 2.

• I can draft the SharpRank outreach email when ready
• Timeline: 1 call + 1 NDA + 1 week of testing

Long-term (month 6): defer Path 3 until we have user + capital scale.

• Not worth the legal + security risk pre-revenue

Ready to ship Path 1 + Path 4 as soon as we're through the Odds API + payment plumbing. Awaiting your nod on sequencing.