By Jane Hales, Managing Partner at Sapio Research
Creating your own original research is one of the most effective ways of standing out from the crowd in both PR and marketing activity. Surveying the opinions and experiences of CISOs and other IT and security decision makers can lend weight to your messaging and help create eye catching headlines.
Here are five top tips for creating a successful research campaign.
1- Know the answers you’re looking for
It is important to have a theory and expected set of results that you want to test in any survey. This makes crafting the survey much easier and helps to ensure that the questions logically flow with a clear structure throughout. It will also mean that you will have a sense of the news headlines and stories that you can create in advance from the survey.
. By starting with desired outputs, we can easily test whether the questions are likely to lead to that specific headline or model (depending on the actual data of course).
The questions should support the story of the headline, and by being clever we can make sure we have back-ups for alternative headers which will still provide a great narrative, even if you don’t quite get the one you were hoping for. However, remember also not to make the questions too obvious or self-serving.
2 – Use closed questions
Wherever possible, try to use closed questions that have a list of ready-made, multiple choice responses or coded answers that respondents can tick. This limits the number of options available to respondents but creates a much more focused set of results.
By contrast, open questions lead to lots of uncategorised data that can be difficult to convert to stats or headlines. That said, the additional freedom of open questions means they are useful for providing quotes that can be used to back up the raw numbers.
In order to write relevant and comprehensive closed questions, you need to be equipped with a strong understanding of the market and issues involved. This is particularly important in a complex and technical industry like cyber security.
3 – ‘Other, please specify’ is your friend
Sometimes it’s just not possible to know all of the possible answers to a question. If you you’re in doubt about whether your answer list is complete, then include an ‘other’ option. This is particularly useful if you are looking to understand the impact of a problem or list of information sources, for example.
Using the ‘other, please specify’ can also be a useful way to road test a survey to see whether you have captured the range of answers to closed questions fully. Use a small number of respondents to test the survey before committing to a full release.
4 – Get more impactful stats
It’s important to consider how responses in your survey will translate into numeric data – known as the code frame. The longer the code frame, the more options respondents will be able to pick from, and the greater the resulting spread of data will be.
If you want to increase your chances of getting one or two impactful stats, then using shorter or more consolidated codes frames is the best option. Long code frames are the equivalent of ‘too much choice’, and results may be too varied to single out any one option.
Another way of getting higher percentages is to use multiple response questions (i.e. tick all that apply); this means that people can check lots of different answers, resulting in higher percentages for each code.
5 – Avoid ranking questions – make things easy
To make it as easy as possible for respondents, try to avoid ranking questions (e.g. what’s your 1st, 2nd, 3rd choice). If a significant amount of mental effort is required to order every option, it becomes time consuming and those who find it too difficult will just choose randomly so they can get through to the next question, resulting in inaccurate data. If you must use ranking questions, try to limit the number of options (codes) to between three and five.
Try to use a rating question or Likert scale (e.g. marks out of five or 10, strongly agree, disagree etc.). This not only makes the questions easier to answer, but also provides more sensitivity in the responses, whilst still allowing them to be reported as ‘first’ or ‘most preferred’. The sensitivity of the data can then not only be used for PR research, but also more complex analysis.
Watch this space for five more top tips from Sapio Research soon. You can also learn more about them here.
In the meantime, check out the below episode of Cyber Beat, the éclat Marketing Podcast for more advice from Jane, including how to maximise results across PR and marketing activity.