Real World Regression R & p-value

I am trying to figure out whether our customer support has an impact on tickets opened by customers. Our employees should contact customers to avoid that a user will open a ticket.

The data is quite accurate. I am plotting the (pro-active) contacts per day, the opened tickets per day and I am using a linear fit for both. Both r² values are around 15% and the p-values are pretty bad as well (way above 5%).

I wonder if I am doing the right thing, if I want to figure out whether there is a relationship between both or if there are better methods (lags etc.?)? Also I wonder if I even should worry about p-values and r², wouldn't a small change over time be more realistic in a real world use case?