The cheapest version: surveys are indicators, not explanations. A good start for minimal cost. One must also consider survey fatigue: after 11 questions, responses often become less thoughtful, which biases even superficial data.