The most important skill for any research or insights professional is being able to analyze research data and produce business-critical insights in a timely manner. Unsurprisingly, those in research roles spend more time on generating insights (i.e., analyzing, interpreting, charting, and / or reporting data) than on any other activity (see GRIT Report 2017 Q3-Q4). Hence, anything that allows research professionals to increase the rate at which they turn data into insight enables them to add significantly more value to their businesses. A research data warehouse improves efficiency and has a game-changing impact on the work of researchers.
An Automotive Example
Before we explore how a data warehouse increases the speed of producing insights, let’s look at some recent data from a handful of mTAB’s automotive clients to provide some context.
This table shows that, on average, these automotive OEMs purchase research data from over nine vendors. While having many sources of research gives a holistic view of important measures such as brand health, customer loyalty, and buyer / rejector behavior, navigating and analyzing data from so many disparate sources would give even the most advanced researcher a head ache! A research data warehouse solves the challenges that come from data from so many sources in two key ways.
A Research Data Warehouse Improves Efficiency by Reducing the Time it Takes to Find Relevant Research
The first key efficiency driver of a research data warehouse is giving researchers instant access to decades of research from any number of different research providers. Having all of an organization’s research data in one place allows researchers to search across hundreds of studies and decades of research in seconds to identify the questions that are most pertinent to her analysis.
Alternatively, if we use one of our automotive clients as an example, when doing analysis on how product quality affects buyer / rejector behavior and brand image, a researcher at OEM 1 would have to log into JD Power’s data portal to access the company’s quality study, Strategic Vision’s portal to access their buyer / rejector study, GfK’s portal to access their brand tracking study, etc. Additionally, the researcher would have to manually comb through the studies to find the specific questions that have the data that she might be looking for. A centralized research data warehouse transforms this task of accessing pertinent data from a manual process that could quite literally take hours or even days to a simple effort of typing keywords into a search bar that takes seconds.
A Research Data Warehouse Improves Efficiency by Reducing the Time it Takes to Perform Standard Analyses
The second key way a research data warehouse improves efficiency is by streamlining analysis. Again, using our automotive example, a researcher at OEM 1 would have to learn up to 14 different tools (one for each data vendor) to deep dive into all the research data and perform standard analyses such as cross tabulations and statistical testing. Learning each new tool is a time consuming and often frustrating process.
Not only does a data warehouse allow users to realize efficiencies by reducing the number of tools they use, it also allows them to standardize crucial analyses and ways of looking at data:
- Segmentation: Using a single data warehousing tool, researchers are able to build a customer or product segmentation and apply it to all of the databases in their warehouse, rather than having to re-build the same segmentation for each study. Many of our clients store hundreds of studies in their mTAB data warehouses, so the efficiencies gained by building and applying a standard segmentation to all studies are immense.
- Trending: A data warehouse allows companies to store all of their research data in a standard format, enabling researchers to ensure that question and response formats are consistent across studies (e.g., from wave to wave or market to market). Thus, researchers can seamlessly trend and / or combine data across time, markets, and products without having to manually map responses or question labels.
- Building Custom Variables: While performing analysis many researchers like to combine questions or responses to build custom variables (e.g., combine gender, age, and education variables to create a new “Young Educated Females” variable or re-bucket responses to an age question to create a generation variable with responses of “Baby Boomers”, “Generation X”, “Millennials”, and “Generation Z”) to focus their results. Much like with segmentation, a centralized data warehouse allows researchers to create these custom variables once and apply them to all of the studies in the warehouse.
Given the growing importance of instant insights and doing more with stagnant research budgets, anything that can increase research analysts’ efficiency in producing a holistic story from research data is worth its weight in gold. A research data warehouse improves efficiency by consolidating research tools, enabling researchers to instantly identify pertinent data, and standardizing crucial analyses.