Three questions stand out. The first is already supported by queries I ran. The second and third depend on the data and are only partly checked so far. ## 1. Is funding accelerating or slowing, and is the market consolidating into fewer, larger deals? Deal counts fall every year while the typical disclosed ticket grows sharply (yearly counts and medians in the rendered result above): - **Deal count:** 935 in 2015, 993 in 2016, 687 in 2017, 310 in 2018 and 111 in 2019. - **Median disclosed ticket:** about $1.5M in 2015 and $1.0M in 2016, then $2.25M in 2017, $4M in 2018 and $12M in 2019. - **Total disclosed USD:** about $8.6B in 2015, $10.4B in 2017, $5.1B in 2018 and $9.7B in 2019. That points to fewer but much bigger deals rather than a simple collapse in activity. It would help founders and investors benchmark how much to raise and when. Three caveats: - Only 2,066 of 3,044 amounts are numeric, so about a third of deals have no usable amount. - The 2016 total came back as `nan`, which means at least one amount is malformed. Cleaning that value is needed before 2016 totals can be trusted. - Part of the decline in deal counts could be falling data coverage rather than the market. 2020 has only 7 deals, and some dates are malformed (one parses as year "0015"). ## 2. Which industry verticals and cities should investors or founders prioritize? The dataset has 822 distinct verticals and 115 cities, which is enough to rank sectors and hubs by deal volume, ticket size and trend. I haven't run those rankings yet. The vertical labels need cleaning first. The number of distinct verticals per year drops from 724 in 2015 to 19 in 2016 and 21 in 2017. That looks like a change in how verticals were recorded, not a real change in the market. A useful analysis would first group the free-text labels into a handful of sectors, then compare volume and median ticket by sector and city over time. ## 3. What do funding paths look like, and who funds whom? There are 3,044 deals across 2,459 distinct startups, so many startups appear more than once. Every deal has an investor name recorded. The data has 57 distinct round types, so you could ask: - How often do startups progress from seed to Series A and beyond? - Which round type is typical for each sector? - Which investors repeatedly back a given stage, sector or city? I haven't computed these yet. Round-type labels and the multi-investor text field will need standardizing first. I can run questions 2 or 3 next if you want one of them dug into.