DIY research do’s and don’ts

We are often asked what we do that free survey or DIY software can’t do. I could go on and on about this, but since DIY research is here to stay, I thought it better to approach this subject from a different perspective and talk about the ‘do’s and don’ts’ of DIY research. So here goes…

Do…research!
Sounds obvious, but generally any research is better than no research. Asking questions of your customers or clients is rarely a bad thing (although it can be misleading if it’s not done correctly!). It shows you’re being proactive and are aiming to improve your offering, experience or overall brand.  In the main, people will respect you for trying.

Don’t…rush it.
Invest time, think about what you want to achieve from the process and use the research as a vehicle to get you there. Plan a timescale and try and stick to it. Working with a research agency can minimise your time investment significantly.

Do…clean your data!
This is one thing I can’t stress enough. Data is everywhere, it’s a hot topic. Manage it well and you can maximise its potential – mismanage it and you can be left red faced. If you’re conducting online research, make sure you’re sending the right messages to the right people and if you’re using their name anywhere, make sure that’s right too!

Don’t…ask leading questions.
Try to take a balanced view, try not to lead people in to saying what you want to hear, test your survey with people outside your organisation – gain perspective. Another benefit of working with an external agency is that we are impartial and can be positioned as ‘Independent researchers’ – which gives respondents a greater confidence that their responses aren’t going to be manipulated to fit a pre-agreed agenda or impact on their future relationships with you.

Do…consider survey flow and routing.
All of our surveys follow tried and tested methods of when to ask particular questions – we don’t rush in with KPIs at the start and we don’t show respondents questions that aren’t relevant to them. An example of this would be asking under 17s if they have a driving licence…you can guess the answer you’ll get.

Don’t…offer an incentive that could bias your responses.
If you offer something that only your loyal customers would want then the likelihood is that only your loyal customers will respond. A neutral incentive will give a better representation of your customer’s true feelings.

Do…brand your surveys.
The quality of your questionnaires is a direct reflection of your brand, so make sure you’re proud of it!

Don’t…spam people.
One or two reminders on a standard email survey is best. Any more and you risk annoying your customers and potentially leaving a detrimental feeling toward your brand.

Do…consider the limitations of free software.
The likelihood is that you won’t be able to cross-analyse the data within the software. This means exporting the data and working through it manually. If you’re a data wiz that won’t be a problem, but if pivot tables and filters aren’t your thing then playing with raw data is best avoided. The benefit of professional software is that it reduces the likelihood of ‘human error’. It also provides additional analytics which increase understanding and avoid misleading interpretation, e.g. significant difference testing which compares sample subsets to show where a percentage difference is or is not significant.

Don’t…be afraid to ask for help.
Many agencies, like us, are happy to lend a hand even if it is only 10 minutes advise over the phone. We understand that budgets can restrict what you can do and we tailor our suggestions to meet your needs and funds.  We are only ever a phone call away to answer questions.

Hopefully this post has helped, whether you’re a first timer or a seasoned researcher. As always, I’d love to hear your feedback or any questions you may have that you think we could answer.

Posted by Jonathan