Thread by @ryan_caldbeck

1/ Randomly thinking about the amount of data that @Lever has. Who is interviewing and what companies think of those candidates. But that data gets super interesting over time and when combined with other data sets.

[I’m not an investor/advisor in Lever and have no affiliation]

2/ First who is interviewing: one of biggest pain points in the recruiting is the drop in conversion between outreach and engagement. Narrowing in quickly on who is interviewing- not who says they are interviewing (i.e. Hired/Indeed), but who is actually interviewing.

3/ What co's think of candidates- Text data and the Lever scores (1-4). Some fun questions:
-Which interviewers feedback at a given co. are most predictive of hiring (or success post close)
-Optimal # of interviewers
-More predictive of the hire – score or long-form review?

4/ When combined with @LinkedIn data:
-% of offers that were accepted
-% of top candidates that joined the co. (perhaps measure of CEO ability to recruit great talent, attractiveness of co)
-connection between who interviewed and whether someone accepted

5/ Data across co's:
Because most candidates are tied w/ LinkedIn (or even those that aren’t, the Entity Resolution shouldn’t be that tough), look at how candidates score across co's. Engineers that always score 4 (of 4) on the Lever scale across all interviews……interesting?

6/ Or top candidates- when they are interviewing at multiple companies, where do they end up? A more accurate measure of talent (or not?) than the LinkedIn data that every VC firm on the planet scrapes.

7/ Clearly some ToS issues and privacy issues w/ using the data (the candidates don’t elect in when they are reviewed as an example). But IF @Lever could figure that out, data seems super powerful.