PHPMem v2.0.1

Version
1.6.45
Uptime
17 days 8 seconds

Memory

Total
512MB
Used
12,72MB (2.48%)
Free
499,28MB

Keys

Current
14 060
Total (since start)
40 994
Evictions
0
Reclaimed
760
Expired Unfetched
0
Evicted Unfetched
0

Connections

Current
15 / 1 024 max
Total
193 883
Rejected
0
llm:c9893990ea1192a6d9875389bef7f8acaa6371ecfeb44074c9590b6cd4225dc2
TTL 6 days 3 hours 21 minutes 19 seconds Size 2,66KB Export
Edit
**There is a time dimension, but it is the month of announcement, not a date column.** Each table is one month (Jan–May 2022). The `Founded` year is a second time axis, but it describes company age, not when the deal happened. The notable breaks are an April spike in dollar value and a May collapse in both volume and data quality. **1. April 2022: a huge spike in dollar value, driven by outliers** - Parsed USD totals were about $4.10B in January, $3.49B in February and $2.75B in March. April jumped to about $45.08B, roughly 11 times January. - A few giant rows drive this. The largest is "Accel" at $18.3B (Financial services, stage "undisclosed"), then Ola at $5.0B, Mahindra Logistics at $2.5B, Verse Innovation at $1.7B and ShareChat at $1.4B. - Accel is normally an investor, not a startup, so that $18.3B row is probably a data-entry or misclassification error. It alone is about 41% of April's total. - The April median round rose to $30M, from $5.45M in January, $6M in February and $12M in March. So April's typical round did grow, not just the outliers. The median had already risen in March, which looks like a step up in deal size. - April's average ($564M) is not a meaningful measure because of those outliers. Use the median. **2. May 2022: a sharp drop in volume** - Row counts fell from 115 in January to 96, 98 and 95, then to 42 in May, a 56% drop from April. - Dollar-denominated rounds fell from 80 to 20, and the May total was only about $299M, with a median of $5.25M. - Part of this is likely an incomplete month, since the data runs only to May. The card also says some May rows are misaligned or blank, so treat it as a data-completeness break rather than a true market collapse. **3. A steady rise in undisclosed or non-dollar amounts, a quality break** - Rounds without a parseable "$" amount rose from 7 of 115 (6%) in January to 10 of 96 (10%), 19 of 98 (19%) and 15 of 95 (16%). In May it was 22 of 42 (52%). - These rows are excluded from the USD totals above. The totals therefore understate May most heavily. - Some values are not amounts at all. The January "max" is the text "Vaibhav Anant" (a founder name, in the Amount column), and May contains a bare "6400000" with no "$" sign. This points to misaligned rows, mainly in January and May. **Bottom line:** January to March is stable, with 96–115 deals and $2.7–4.1B per month. April is a one-month value spike, partly an Accel data error. May is a volume and quality drop that is probably partial data. Any trend analysis should drop or flag the April outliers and the May misaligned rows first.