Please have a look at these two questions.

1) How satisfied are you with Apple's delivery and payment options?

2 ) Would you buy this improved version of Twitter?

If you have been using questions like these in your survey, you have been missing out on valuable insights. I'll tell you how the framing of the questions can mess up your results, and how to fix your data:

4 Don'ts for survey design

Avoid leading questions - use neutral language. A leading questions' wording influences the respondent to answer in a particular way. This is the fallacy of example 2.

Avoid agree/disagree statements - use a rating scale. You can use agreement statements, but make sure to use a rating scale. This is because respondents tend to agree out of politeness (to learn more, google "acquiescence bias").

Avoid “double-barreled” questions - isolate them. Double-barreled questions ask respondents multiple things at once. As respondents can feel different about each item, interpreting the data is impossible. Instead, isolate them. This is the fallacy of example 1.

Avoid hypothetical questions - focus on the current experience. People can't accurately predict how they would act in a hypothetical situation. Everybody likes more features - but how many would use them? Instead, ask about the current experience to find pain points (and solve those).