More thoughts on Multiple Source Analysis
I’m sure you don’t need to be told that there’s quite a lot of data about at the moment. Apropos this, we blogged recently about the dramatic increase in projects requiring us at Bonamy Finch to pull together data from a number of sources. We call this type of project Multiple Source Analysis, in lieu of any other description that accurately defines them: ‘Big Data Analytics’, ‘Predictive Analytics’ and ‘Data Science’ all seem too far removed from the interpretation that all good researchers place on their analysis; while ‘Business Analytics’ feels like a catch-all, management consulting-style term.
A few examples…
Multiple Source Analysis, (or MSA as we have started to refer to it as), whilst admittedly a bit generic, seems to capture many assorted types of analysis project that come through our doors. Recent Bonamy Finch examples include:
…Using meteorological data to explain shifts in sports participation over time
…Building a customer segmentation using a combination of survey data, behavioural databases and Experian data
…Overlaying geographic boundary tags onto customer-level information to plot customer incidences on Google Maps
…Modelling target audience reach using brand KPIs, competitor channel spend
…Merging SKU-level sales and click-through data with TURF analysis from a retail consumer survey, to rationalise a proliferation of new products
…Supplementing expected profit figures to individual drinking occasion data, to size opportunity areas
…Triangulating survey volumetric forecasts with government and industry data in the financial services market, to provide more accurate results
…Fusing disparate ad-hoc research datasets to community panel data, to create a single consumer source of insights
Whoever you speak to in this business, there is a consensus that Multiple Source Analysis will become even more prevalent in the next few years. Clients will increasingly look to agencies to synthesise multiple datasets throughout the research process – with a number of benefits:
- A thorough materials review up front, alongside closely aligned datasets, will allow us to identify and fill knowledge gaps efficiently and quickly
- Re-using (or, in many cases, just using) previously-collected data, in a cost-efficient way
- Corroborating insights, and joining up the dots, before they hit the client’s inbox will mean we hit the mark more often, and sooner
- Additional commercial context this gives will increasingly allow researchers and analysts a voice in the boardroom
The latest Greenbook GRIT Report (link) confirms “that the ability to understand, analyse, and use various types of data stands out as the most necessary skill” in creating successful future researchers. It is seen as more important than storytelling, visualisation, business impact, and even MSA’s slightly geekier cousin, Data Science.
In the next few weeks we will be distributing a short introductory overview on Multiple Source Analysis. This will outline the main distinction between predictive and exploratory models, and the different types of projects within each of these two main branches (with case studies to make things really clear). Reading the overview won’t make you an expert – but it should help you to start having ideas and conversations about how you might get more impact from the data you, your colleagues or your clients already have.
If you are not already on our mailing list and would like to receive the overview on Multiple Source Analysis, then please just send an email asking to join to email@example.com