The only thing worse than no data at all is bad or “unclean” data. At PortMA, we take the integrity of our data very seriously. If you are unsure about what steps to take to ensure that your data is clean, look no further. I am here to help.
Below are the steps we take at PortMA to clean our data.
1. Preventing "straight-lining" or "speeding"
We monitor the length of time it takes someone to take an on-site, consumer-intercept survey. All event marketing surveys are different. A five question survey won’t take as long as a ten question survey, so there isn’t a universal baseline. We take each survey on a case-by-case basis and determine the average amount of time it should take to administer it. At that point, we remove any responses that appear to be too short, or too long.
2. Ensuring survey completeness
We require a minimum number of questions to be answered. Having robust event marketing data is important, but including responses that only have one or two questions answered can artificially expand margins of error. There is not a minimum universal number here. We use a percentage – 40% to 50% of questions answered – as a rule.
3. Verifying survey responses make logical sense
We clean event marketing data after it has been collected. This does not mean that we tamper with the data or change responses to say what we want them to. We “correct” quirky responses. For example, a consumer says they are having their first experience with a product at an event but then report they currently buy the product five times a month (this is hypothetical). That doesn’t make sense.
In this situation we have two options. Either we modify the first question to indicate the consumer is a current customer, or we could change the number of purchases per month to zero in an effort to make the responses copacetic.
Please note, we take data correction very seriously and it is not done on a whim. The client and senior management are involved in each decision of back-cleaning data.
4. Reviewing survey response logic
We require questions that “break” the consumer survey. In other words, we use logic to direct people on different paths in a survey based on their response to an earlier question. This means that if the qualifying question isn’t answered, then the survey will be broken. As a result, the survey will not know how to redirect the respondent.
It is important that you are careful with questions that you require to be answered. Make sure that you only do it only with questions when absolutely necessary. You don’t want to make taking a survey hard on the consumer.
In conclusion, ensuring the integrity of our event marketing data is one of the most important things we do at PortMA. By following the steps above, we ensure our data is of the highest standards.