如果出现新客户，如何从给定客户最接近的旧记录中获取日期

Question

I am dealing with policy records table inside Google BigQuery. My table has three columns: INDVDL_ID, POLICY_ID and CNV_DT. Every unique client is identified with INDVDL_ID and may have one or more records in the table depending on how many policies he has. Each policy has it's id (POLICY_ID) and the date that it was bought (CNV_DT). So the code for pulling the data that I am using looks like that:

SELECT INDVDL_ID, POLICY_ID, CNV_DT
FROM `Policy_table.TP_Policies.policy_20191023

What I have:

 <table><tbody><tr><th>INDVDL_ID</th><th>POLICY_ID</th><th>CNV_DT</th></tr><tr><td>1</td><td>1</td><td>2008-01-01</td></tr><tr><td>1</td><td>2</td><td>2008-04-31</td></tr><tr><td>1</td><td>3</td><td>2008-12-23</td></tr><tr><td>3</td><td>4</td><td>2009-08-19</td></tr><tr><td>2</td><td>5</td><td>2010-06-12</td></tr><tr><td>2</td><td>6</td><td>2011-11-12</td></tr></tbody></table>

What I would like to pull is table, where for every additional policy that customer has bought I can have a CNV_DT of his prior purchase.

What I would like to have:

 <table><tbody><tr><th>INDVDL_ID</th><th>POLICY_ID</th><th>CNV_DT</th><th>PRIOR_CNV_DT</th></tr><tr><td>1</td><td>1</td><td>2008-01-01</td><td> </td></tr><tr><td>1</td><td>2</td><td>2008-04-31</td><td>2008-01-01</td></tr><tr><td>1</td><td>3</td><td>2008-12-23</td><td>2008-04-31</td></tr><tr><td>3</td><td>4</td><td>2009-08-19</td><td> </td></tr><tr><td>2</td><td>5</td><td>2010-06-12</td><td> </td></tr><tr><td>2</td><td>6</td><td>2011-11-12</td><td>2010-06-12</td></tr></tbody></table>

Answer 1

You seem to want lag() :

SELECT INDVDL_ID, POLICY_ID, CNV_DT,
       LAG(CNV_DT) OVER (PARTITION BY INDVDL_ID ORDER BY CNV_DT) as PRIOR_CNV_DT
FROM `Policy_table.TP_Policies.policy_20191023;