Benchmarking

Benchmarking is an important use. There are typically two types of benchmarks: longitudinal and cross-sectional. On our platform, several features are related to benchmarking. They are: targeting, data overlays, date range segmentation, and other segmentations.

Longitudinal Benchmarking

Very often, for a need that requires collecting the same kind of data on an ongoing basis, a company may keep the same survey open for a long period of time. When insights accumulate as data get collection continues, there are two ways to cut and dice the data temporally.

  • The first is to create date-range based segmentation. For example, the segments can be for “2015Q2”, “2015Q3” or “2015Q4”, respectively, and named accordingly.
  • The second is to use targeting. For example, to continuously replace the survey URL being used. Each target can be named accordingly too. For an event venue, these targets can be named by events. For a corporate, these targets can be named by project stage.

The first way has very low logistical overhead. The second way trades higher overhead for better specificity. Both ways are can be performed via DIY entirely. If you have questions, please contact user support.

Cross-Sectional Benchmarking

You may need to compare feedback from different sections of the survey audience. For example, different divisions, job functions, age groups, seniority levels, etc. There are two ways to cut and dice the data temporally.

- Segmenting on responses to the questions in the survey. For example, by all kinds of organizational demographics information.

- The second is to use targeting and/or data overlays. In this case, you don’t have to ask for literal answers but you have the basic information that is required.

Like with longitudinal benchmarking, the first way for cross-sectional segmentation has very lower logistical overhead. The second way trades higher overhead for better specificity and more information that can define a more comprehensive context. Both ways are can be performed via DIY entirely. If you have questions, please contact user support.

Benchmarking Across Multiple Surveys

There is sometimes a situation where the survey administrator runs a few surveys (over a period of time), where the scopes of the surveys are largely different, but there is a core subset in common. This core subset can serve as the basis of creating some benchmarks. This need is often an agency need, although sometimes an end-user (corporate) may have this need as well.

There are two ways of performing benchmarking across surveys.

  • Inventory Survey. One way is to keep an inventory of surveys. When running a new survey, just copy from an inventory survey, rebrand it, customize the content, and run it. For benchmarking, this is appropriate when the set of benchmarking questions are still being refined. The survey administrator can also develop and evolve the methodology that produces a set of benchmark, each represented as a “score”. These scores can be created on a per-survey basis, and done outside our platform.
  • Survey Overlay. The other way is to use an advanced feature called Survey Overlay. It is related to data overlay, only that our user support needs to set that up for you. Please contact user support.

Survey Overlay

To use survey overlay in benchmarking, logistically you’ll need to keep a long-running benchmarking survey. That’s appropriate when the question set has been finalized and is generally applicable to many scenarios.

We refer to this benchmarking survey as the “B” survey. The varying part of your surveys are referred to as the “A” surveys. You’ll have many survey As and they are all different (e.g. A1 for Pfizer, A2 for Merck, A3 for Novartis, …). There is only one survey B (with B being the benchmarking one).

Survey B has many survey targets, one per each survey A. The survey for Pfizer is then A1:B, for Merck is then A2:B. A1, A2, A3 … are one-time use, turn on and then turn off. B is that long-running survey. With this setup, in survey B, you can check per client data by picking the target, or check a particular time-period by doing a date-range segmentation.

When the B survey is invoked via different survey target, the appearance of the B survey can vary according to the target used. For example, the target corresponding to A1 (for Pfizer) can trigger the B survey to bear the same survey theme used for survey A1. There is no limit on how many targets survey B can have.

The primary benefit of doing benchmarking by way of survey overlay is that all of the data analytics, including the Action Priorities Grid, are provided automatically. Thus saving a lot of time for the survey administrator.

All enterprise tier subscriptions include survey overlay as an advanced feature.