Updating NaN column (with string) based on multiple columns

The question is very similar to Updating column in a dataframe based on multiple columns

I have the following dataset:

CustomerID  TypeofContact   Occupation  Gender  MaritalStatus
0   200000  Self Enquiry    Salaried    Female  Single
1   200001  Company Invited Salaried    Male    Divorced
2   200002  Self Enquiry    Free Lancer Male    Single
3   200003  Company Invited Salaried    Female  Divorced
4   200004  Self Enquiry    Small Business  Male    Divorced
5   200005  Company Invited Salaried    Male    Single
6   200006  Self Enquiry    Small Business  Female  Divorced
7   200007  Self Enquiry    Salaried    Male    Married
8   200008  NaN             Salaried    Male    Single
9   200009  Self Enquiry    Small Business  Male    Divorced
10  200010  Self Enquiry    Small Business  Male    Divorced
11  200011  NaN             Salaried    Female  Single
12  200012  Self Enquiry    Small Business  Male    Married
13  200013  NaN             Small Business  Male    Married
14  200014  Self Enquiry    Salaried    Male    Single

I am looking to update the NaN in TypeofContact column to the first record that satisfies the the condition - combination of not null Occupation, Gender & Marital Status with the null record's combination of Occupation, Gender & Marital Status

Example:

CustomerID with 200014 can satisfy the CustomerID 200008 that has TypeofContact as NaN as both sets have the same Occupation, Gender & Marital Status

Same is the case with 200013 and 200012

CustomerID  TypeofContact   Occupation  Gender  MaritalStatus
0   200000  Self Enquiry    Salaried    Female  Single
1   200001  Company Invited Salaried    Male    Divorced
2   200002  Self Enquiry    Free Lancer Male    Single
3   200003  Company Invited Salaried    Female  Divorced
4   200004  Self Enquiry    Small Business  Male    Divorced
5   200005  Company Invited Salaried    Male    Single
6   200006  Self Enquiry    Small Business  Female  Divorced
7   200007  Self Enquiry    Salaried    Male    Married
8   200008  Self Enquiry    Salaried    Male    Single
9   200009  Self Enquiry    Small Business  Male    Divorced
10  200010  Self Enquiry    Small Business  Male    Divorced
11  200011  Self Enquiry    Salaried    Female  Single
12  200012  Self Enquiry    Small Business  Male    Married
13  200013  Self Enquiry    Small Business  Male    Married
14  200014  Self Enquiry    Salaried    Male    Single

I was able to create another dataframe with not null's, loop through it and update the original dataframe using CustomerID identifier.

What would be an efficient way to accomplish this?

Thanks.