Our difference starts with our heritage and our pedigree. Many of the team at Investment Trends were pioneers in the online research industry, having started their work in the area in the mid-1990s. And while there are many ways to run online research today, building a correct, representative sample of niche segments or complex industries is a very difficult undertaking.
The difference between everyday research and deep insights is a function of every step of our process. From our survey design and data collection processes, perfected over our 12 years of researching, to how we treat data once it’s collected, to the industry expert overlay we apply, the Investment Trends process is engineered to go deep – where real value can be unlocked for our clients.
The data collection and sample weighting processes used in quantitative research are absolutely integral to the quality of the outputs. At Investment Trends we do an enormous amount of work in both these areas, and we refuse to move to the next stage of our research until we’re convinced we’ve got them right.
The output of our research recognises the complexity of the issues the industry faces and is delivered in three parts:
- Clients receive a deep data research pack that provides every bit of data we have extracted – running to hundreds of pages. This data set is an important underpinning of our service, but while its incredibly interesting and valuable, it is only the first step in how we deliver value.
- To help clients turn this data into actionable deep insights our Research Director prepares a pack for presentation to the senior members of our client’s team. This presentation is bespoke for each client and is designed to facilitate extensive discussion, ensuring the team leaves with clarity on what the research says, what it means for the industry, what it means for our client, and what decisions they can make to improve their competitiveness in the short, medium and long terms.
- Our clients inevitably ask if we can cut the data we deliver in all sorts of creative and insightful ways. All our research is delivered with the assumption that we will provide further data cuts post presentation to support our clients, because we deliver deep insights – not reports.
At the statistical level we utilise numerous forms of advanced multivariate analyses, including:
- Principal Component Analysis
- Discriminate Analysis
- Factor Analysis
- Cluster Analysis (hierarchical and k-means clusters)
- Two step cluster models
- Logistic Regression Models
- Binary Logistic Models
- Multinomial Logistic Models
- Analysis of Variance (Two way classifications model)
- Factorial Design
- Choice modelling (multinomial and binomial choice modelling)
- Correspondence analysis
- Optimal scaling
- Variance component analysis
- Auto-regressive models
- Exponential smoothing
- LINEX modelling
- Predictive clustering model