Notes from Rachel Petersen's discussions on surveying audience

pcampdc1.txt

SurveyMonkey, Cvent great surveying resources

How do I improve survey responses?

1. Work backwards
	- what question am I trying to answer?
	- what data do I need to answer the question?
	- what type of survey should I use to collect this data?
	- to whom should I ask the question

If the idea is easy to articulate and identify, you can easily put in a survey. Articulation allows you to quantify. 

Surveys are perfect for concrete, well-formed, filterable ideas with a targetable segment. If you're fuzzier, you need a focus group where you can discuss and flesh out your idea. 

2. Keep your survey simple
	- use customer language
	- ask clear questions and be concise
	- limit time to complete the survey by limiting the number of questions

Only ask for data that you don't have or you can't get. If you know what the revenue of your customers is, don't ask them again.

3. Watch out for bias
	- limit leading questions and language
	- randomize questions when possible
	- open ended questions
	- include an "n/a" or "don't know" option

4. Incentivize survey response
	- be up front about how survey responses will be used
		- a lot of times, the incentive can be as simple as telling customers how you're going to use their data
	- offer a prize!
		- but know your audience and what they want
	- donate to charity for each survey completion

Use your branding! Make your survey look legitimate. 

If you're donating to charity, a government official or an individual that might be foreign-based, get your lawyers involved! There may be foreign bribery concerns.

The line for statistical significance of a survey response is extremely low. You have a statistically significant answer with 1000 responses from a survey of 100,000 customers. 

5. Data!
	- visual representations of data - select the right type of chart
	- relevancy
	- sample size
	- combine with other data sets if possible