Efficient vs Inefficient Market Research Data Processing
I am sometimes shocked to see how inefficiently s…Read more
There are a handful of software programs that serve the research market so that cross tabulations (or crosstabs as they are often called) can be scripted – Quantum, MRDCL and Merlin are three of a small bunch. Cross tabulations are used extensively in market research to analyse one question or variable by another question, variable or set of questions.
There is a distinct split between cross tabulation packages that are driven by a script and those which are driven by a user interface (non-scripted). They both have strengths and weaknesses, of course. And, of course, some are better than others.
Scripting languages offer two main positives – complex requirements can be programmed and repetitive requirements can be handled efficiently without the need to copy and paste or replicate steps. The problem with scripting languages is that they need skilled / more expensive staff to use the software and easy tasks such as entering question and response texts may need to be handled by overqualified staff whereas less skilled staff can handle such tasks just as well.
The biggest benefit of non-scripted tabulation, in most cases, is the ease of use. Users do not typically need a lot of training to get started, although some packages do have a lot of options that can take some time to become familiar with. They are ideal for handling many ‘standard’ market research surveys and most online surveys. They can become slow to use if requirements are repetitive or difficult. Typically, these types of programs only offer one or two solutions to something more difficult than straightforward tabulations which means that smart solutions may not be available. There is always a limit to what actually can be produced and compromises on output may be necessary.
We firmly believe MRDCL is the best scripted tabulation software in the world. But, that doesn’t mean you should use it! There are several reasons not to use MRDCL or any scripted tabulation software.
Firstly, the users will need to be regular users, probably spending at least half of their time using this type of software at some time in their career. They will need to be trained not just to learn the basics but, to be of real valuable, they would need to understand a sufficient amount of the more advanced techniques. This means investment and training. If the work being carried out is generally straightforward, there is little reason to use over-skilled staff. Indeed, companies that only have one or two complex projects per year would usually be much better served by using a simpler non-scripted tabulation system and subcontracting the more complex project. Another reason not to use a scripted software product is cost. MRDCL is more expensive than most non-scripted software products but, of course, scripted software usually offer more.
MRDCL is a good fit when your company has a high volume of work where some or all of the work is either complex, requires repetitive analysis, uses big questionnaires, has large data sets, has complex data sets or, perhaps surprisingly, where you have a lot of similar projects in which many aspects of the project are almost identical. In such cases, templates can easily be built on top of MRDCL to make these easy to manage and run.
MRDCL is different from the other scripting languages. Arguably, its biggest key feature is a technique called EPS, which stands for Excel Productivity Scripting. This allows users to build templates in Excel that MRDCL can read to automate anything that you might want to re-use. You can design the templates more or less exactly as you want them so that MRDCL reads them as input to control what output you get.
Simply, EPS means that skilled staff spend their time doing skilled tasks and less skilled staff spend their time less skilled staff. This can reduce costs, improve project management and increase staff flexibility.
There are numerous examples where clients have implemented systems that offer huge productivity gains using EPS. One example is storing code lists with codes in a spreadsheet denoting which sub-totals each code belongs to. MRDCL can read this and automate output such as ranking results within each sub-total. Something that would laborious to manage in any other scripting language. Another example is using a spreadsheet to choose which types of tables are automatically generated from a series of rating statements – for example, ranked mean score summary, top 2 box analysis, summary of each statement and responses etc. This can all be driven by keywords, texts, codes or any other notation in an Excel spreadsheet - and re-used from project to project. This saves a lot of time and can improve productivity by 50% or more.
BUT, there is another real benefit. Scripts in MRDCL, Quantum or, in fact, any programming language are notoriously complex to take over when someone leaves or is on holiday. When projects are managed through spreadsheets, the time it takes to take over someone else’s work is a fraction of the time it takes to take over a project that is fully scripted. It also means that projects can be shared between skilled and less skilled staff. Basic texts and question specifications can be carried out by junior staff leaving the higher paid skilled staff to focus on things that need their level of skill. This reduces costs substantially and gets over the perennial problem that users of scripted tabulation software find it hard to share projects.
Using these EPS techniques, the right person is doing the right work. Producing cross tabulations becomes just like any other business process where the more skilled person does the parts of the process that require his or her skill while the simpler parts are coordinated and put together by an assistant.
If crosstab work is organised efficiently, tables can be generated far more cost effectively than using non-scripted software solutions. The mix of right staff, using EPS and sufficient training can make a small or large team of people highly effective provided there is a reasonable volume of work that is not just comprised of small, simple projects.
To get the best from MRDCL, these changed work practices can mean re-thinking how your data processing function operates, but the gains are big. I have always believed that less skilled staff do more basic work better than over-skilled staff. It’s kind of human nature!
Changing to MRDCL from another scripting language like Quantum is not difficult. Most users convert fairly quickly, although there is a second phase of learning where it is beneficial to learn the more advanced techniques. This can sometimes mean approaching projects differently from other less flexible scripted tabulation software.
If your company has never used scripted tabulation software, there is a learning curve. Benefits will not arise immediately. You will need someone with the right skillset (isn’t that true of every job?), but you will see continual improvements in productivity as MRDCL is focused on productivity.
We genuinely don’t want unsuccessful MRDCL installations – it’s not for everyone – we have other products if MRDCL is not the right fit. If we believe MRDCL is the right fit for you, we will be pleased to present you with a clear proposal that will explain how you can be a successful user and how you can enjoy significant productivity gains.