Overview
- Many surveys are simple one-off questionnaires. They are programmed, the data is collected and the survey itself is no longer required. Other surveys, of course, are trackers but the desire to change as little as possible between waves makes trackers quite similar to one-off surveys – they just “live” for longer!
- However, there is a third type of survey. Surveys that are separate but require a strong “family resemblance”. For example – brand equity models will always require a core set of questions asked largely in the same way and with identical scales etc. Even the broad wording needs to be kept “to the script” or the model won’t work. Likewise repeat surveys, like concept tests etc, really benefit from consistency. In fact the greater the consistency the better research decisions can be made because the data will be on a common base.
- So – if consistency is a good thing and as much of a “repeat survey” should be kept identical – then manually programming each instance is, at best, a complete waste of resource. Even worse manually programming each one can be error prone and it’s almost impossible to “align” the output data into a suitable benchmarking data warehouse.
Our research automation tools solve these problems for you.
- Survey In a Box – takes your repeat survey and splits it into those elements you want to be consistent and those that need to change. Typically, for instance, the name of the market and the brand set will change but the bulk of the question wording and the scales etc used should not.
- Results In a Box – because Survey In a Box stores data in a structured and consistent way we can present the results consistently as well. This can include automated tables, complex calculations and even fully charted output. Naturally benchmarked comparisons can be provided as a matter of course.
Client Benefits
- Automated reporting and built-in benchmarking
- Extremely fast creation of your repeat surveys (once designed of course). These can – optionally – include our advanced NexGen survey questions. The time taken to build a new instance of a survey is typically 2½ minutes.
- An opportunity to create and lock in client's intellectual property by establishing and fixing a best practice process.
- Considerable labour cost savings
Pricing
- Charges are for survey audit and design, plus scripting of the new survey and design of customisable components within the survey.
- Charged at an hourly rate of £150/hr during design.
- Thereafter project management is £95 / hour
How does it work?
- Assuming you already have a repeat survey that is suitable then we start with a simple discussion around the survey design and which parts of it can be “tokenised”. For example, in the question “How often do you buy Ground Coffee?” we can create a “token” for the market. Our modified question then becomes “How often do you buy #Market_name#?”. Other tokens would typically be brand lists, possibly time between purchases (most people buy Ground Coffee more often than a new TV!) and maybe market attributes.
- Having created a tokenised version of your survey we then programme this into our Survey In a Box system to create a master survey.
- The tokens themselves are turned into a “Wizard”; effectively we create a survey about your survey. The wizard completes the tokens for each instance of your survey (What should I use as the market name, please enter your brand list etc). Complete the wizard and the new version of the survey is ready to launch. You choose whether this is something we do on your behalf or to use your own team.
- Optionally we can automate calculations from the base data. For instance you might ask us to weight brand attribute performance based on the importance each respondent attaches to each attribute. We can even automate your complete model if you have one.
- Of course the repeat survey may need to be part of a larger survey – with extra questions before and/or after. This is fine too.
- Finally, if you want automated output through our optional Results In a Box then you need to tell us what output data, calculations and charts you want to receive. You may also want to attach “meta data” (Market, Country, Client etc) for better filtering of your benchmarking data.
Deliverable
- A functioning system for scripting a survey in flash using a automated wizard
- A functioning system for displaying the structured data collected as agreed tables and reports.
Call our Research Director Greg Ward on +44 (0) 8700 427 427