Frame challenge: this survey does not indicate "what you should be paid".
The question uses phrases that imply a direct and unequal relationship between employee worth and pay, such as "should be getting paid more", "increase salary to the appropriate level", and "underpaid". But the comments on the question reveal that this was simply a statistical review of salaries across the industry, showing what other people in similar job roles make. While this survey shows what you make relative to your peers, it does not reveal what you make relative to your worth - i.e. "what you should be paid".
By definition, half of employees make a below-median salary, but the fact that they make less than average does not by itself indicate that they should be paid any more than they are. A junior or poor-performing employee will have an appropriate salary at the low end of the pay scale. A useless employee isn't worth paying even 50% of the market rate. A superstar employee might be in the top 10% of salaries, and still deserve a pay raise.
While a gap as large as 30% suggests that the employee may well be underpaid relative to their worth, it does not necessarily mean that person could or should command a 30% higher salary for the work they produce. Getting a 30% raise would be a big bump in pay - it's unlikely the employee could achieve that in-house. If they're really worth 30% more than they're being paid, they should find a job elsewhere. But keep in mind that this survey is no indication that this particular person with this particular skill set and experience could actually achieve 30% higher pay anywhere.