In-person focus groups absolutely sound like nonstandard oral communication and direct observation to me. You’re not making statistics with the findings (I hope!) and you’re using a field guide or sample questions to direct the conversation (not a series of exact questions you’re reading like a robot). A rule of thumb is that if it takes a human (as opposed to a printout or web form) to conduct (like a focus group) you’re likely having a nonstandard oral conversation. There’s been some clumsy guidance in the past that made it sound like it might need review, but my recommendation would be to stick with the exceptions outlined in 5 C.F.R. 1320.3(h)(3).
For surveys, I’ll break my answer into two chunks.
- *Can* you do surveys without 9 months+ of review? It depends on what you’re collecting. If it’s general, open-ended questions about experience “How well did we serve you today?” or star ratings “Rate 1–5 stars” then yes (and guidance covers those scenarios). If it’s something you want to use to make statistics to make policy “4 out of 5 small business owners think that policy Y is how we should move forward!” then you definitely need help from your PRA officer.
- *Should* you do online surveys to better understand target audience preferences? In the federal context, I strongly recommend any method other than surveys. What I’ve found is both that you get a weird sample for government online surveys (what weirdo would fill those out!?[me]), and that for the headache and work to get a survey written and approved, you can do incredibly meaningful research with phone calls, emails, interviews, focus groups, etc. Some of the best methods that are alternatives to surveys are outlined here: https://methods.18f.gov/