Yes. The records are funding deals, and four fields group them naturally. Each needs label cleanup first, because the raw categories contain many spelling variants. **1. Funding round type (`InvestmentnType`) is the strongest split.** - Two labels cover almost everything. "Private Equity" has 1,356 deals and "Seed Funding" has 1,355, together about 2,711 of 3,044 rows (roughly 89%). - The long tail is mostly variants of the same few groups: - Seed/Angel: "Seed/ Angel Funding" (60), "Seed / Angel Funding" (47), "Seed\nFunding" (30), "Seed/Angel Funding" (23), "Angel / Seed Funding" (8), "Seed Round" (7), "Seed" (4), and "Seed / Angle Funding" (3). - Debt: "Debt Funding" (25), plus single-row variants such as "Debt", "Debt-Funding" and "Term Loan". - Priced rounds: Series A (24), B (20), C (14), D (12), plus a few Pre-Series A, E and F. - A sensible grouping is Seed/Angel, Private Equity, Priced Series (A–F), Debt, and Other/Unknown. A few rows are blank or "nan". **2. Time (`Date dd/mm/yyyy`) shows the lifecycle of the market.** - Deals by year are 935 in 2015, 993 in 2016, 687 in 2017, 310 in 2018, 111 in 2019 and 7 in 2020. - That is a peak in 2016 and a steep decline afterwards. One row has no date. - Year works as a cohort or era segment. **3. Geography (`City Location`) is concentrated in a few hubs.** - Bangalore (700) and Bengaluru (141) are the same city. Mumbai has 567. - Delhi NCR is split across several labels: New Delhi (421), Gurgaon (287), Noida (92), Gurugram (50) and Delhi (34). - Pune (105), Hyderabad (99) and Chennai (97) follow. - The column has 944 distinct values, which is mostly noise: typos ("Ahemadabad"), non-breaking-space prefixes, and multi-city entries like "Pune / US" or "Bangalore / SFO". 171 rows are "nan" and a few are blank or "N/A". - Normalising to a metro (Bengaluru, Delhi NCR, Mumbai, and so on) gives a usable grouping. **4. Industry vertical (`Industry Vertical`) is useful but needs consolidation.** - Top labels are Consumer Internet (941), Technology (478), Healthcare (70), Finance (62), Logistics (32), Education (24) and Food & Beverage (23). - eCommerce appears as "eCommerce" (186), "ECommerce" (61) and "E-Commerce" (29), about 276 combined. - 171 rows are "nan". - The column has about 900 distinct values, so group it into the top 10–15 verticals plus "Other". **Recommendation:** use round type and year as the primary segments, with metro and consolidated vertical as secondary cuts. The round-type, city and vertical labels need cleaning before any grouping, to merge case, spacing and spelling variants and to treat "nan", blank and "N/A" as one "Unknown" bucket. Several of the queries behind the round-type counts (steps 0 and 3–10) show the same top groups. Amount is stored as text, so any funding-size comparison across these segments would need it parsed to a number first.