Big Sister AI

Sales leaders manage millions in pipeline on gut feel and lagging CRM data. Managers review under 3% of their team's conversations, then decide who gets promoted, coached, or fired. And when humans do review calls, they don't agree with each other. In our own study — 20 meetings, 16 skills, three trained expert raters — inter-rater agreement measured at Krippendorff's α 0.319, far below 0.667, the floor for even tentative conclusions. Sales skill isn't just under-measured; it's unreliably measured. The result: skill gaps and preventable deal losses stay invisible for roughly six months, until the pipeline finally shows it. By then it's too late to coach, correct, or recover. On a typical team, rep win rates range from 2% to 22% and nobody can say why.

Every call, email and CRM record gets an objective 0–100 Sales Score, broken into 18 skills, with every score citing the transcript evidence behind it. Leaders see exactly where revenue leaks; reps see precisely what to change this week. How it's validated: scoring runs on a frontier LLM (currently Anthropic's Fable, model-agnostic by design) against a hand-built rubric — 17 years and 200+ sales teams scored manually before we wrote a line of code. Added as a fourth coder to a three-expert rater panel, our scorer changes group reliability by 0.003 — statistically indistinguishable from adding another trained expert, at 100% coverage instead of 3%. What nobody else has: an appeals court. Any rep can dispute a score; over 10% of interactions are pulled for second-judge review. Every appeal, override and independent re-score becomes a labelled human judgment that refreshes our Golden Set. A referee who merely accepts the AI's score doesn't count as a judgment — that rule is enforced in the data model, so the measurement stays honest as we scale. Live on AWS. Works with Salesforce, HubSpot, Pipedrive and Zoho; calls, Zoom, Teams, Meet and email. Gong and Clari record and forecast. We score. Most of our customers run both. Gong can add a score. It cannot add an appeals court, a hand-built rubric with 17 years of provenance, or cross-company benchmarking — those need an accumulated Golden Set, not a model. Conversation tools analyse calls; we govern the full revenue motion across calls, email and CRM, at rep level, with cited evidence. Built for US B2B teams of 5–30 reps already running a CRM and a notetaker — the segment the enterprise platforms don't serve.

$9,150 MRR (week of 26 July 2026) across 3 paying customers and 46 scored seats — all migrated to V2, which is live and paid for. Two new logos land in August, taking MRR to ~$15,150 (+66%) and cutting single-account concentration from 93% to 56%. De-concentration is exactly what our upcoming round funds. Operating at roughly break-even on a ~$4.3K/month cash base. $360K raised to date; V2 built and shipped on it. Durability: our longest-running customer has paid every month for 30 consecutive months. V1's churn came from manual onboarding overhead — V2's self-serve onboarding removes it, which is why we restarted metrics at the V2 launch rather than carrying V1 numbers forward. Recognition and partners: 1st place, Texas AI Challenge ($100K, Aug 2025). Backed by Capital Factory, Plug and Play, AWS and Microsoft. Anthropic Claude Technology/ISV Partner. Pricing: Solo $30/scored rep/mo · Team $50/scored rep/mo (min 5) · Custom from $5,000/mo. Only reps whose calls get scored pay — managers, admins and viewers are free.

• 
  Ukrainian SalesTech Startup Big Sister AI Receives $100,000 at Texas AI Challenge | Scroll.media

  Scroll.media · Aug 20, 2025

  Ukrainian SalesTech startup Big Sister AI wins $100K at Texas AI Challenge, boosting its $1M early-seed round and expansion in Dallas.

• 
  Creatio Partners with Big Sister AI to Further Transform Sales Team Productivity with AI-Driven Solutions

  SalesTech Star · Jan 24, 2025

  Creatio, a global vendor of a no-code platform to automate workflows and CRM with a maximum degree of freedom, partners with Big Sister AI, an AI-powered sales performance platform designed to improve performance in sales teams by offering AI-driven analytics,

• https://dev.ua/en/news/ukrainskyi-startap-big-sister-ai-1755676448

  dev.ua

• https://www.bizjournals.com/austin/news/2025/08/25/big-sister-ai-wins-texas-ai-challenge-austin.html

  www.bizjournals.com