Articles

Sampling for volume

In market research, market research fieldwork on May 7, 2012 by sdobney

In general, in market research studies the aim of the sample is to represent the market. That is we try and take a list of consumers or customers and ask a random sample of them to take part in the research. In practice, the ‘list’ may be people found on the street, or by door-knocking or from random digit telephone dialling, or from an online panel often with the aim of a 1 in N type sample. However, in markets that follow the 80:20 rule (and that could be most), sampling by volume could be more effective.

In the 80:20 rule (also known as the pareto rule – very common to database analysis) 80% of the volume in the market comes from 20% of the customers. Now if we have a random 1 in N sample based on all people, it would mean that 80% of our sample are responsible for only 20% of purchases. So when we look to the analysis, most of the responses come from low volume buyers and this group dominates over our key group of volume purchasers in our analysis. In other words a majority view from our sample could actually be totally opposite the view of the buying heart of the market.

In practice, these things aren’t so extreme in consumer markets, but it’s something to be aware of (and so check you look for any differences between heavy and light purchasers), but it doesn’t necessarily impact the sampling plan.

In business-to-business and technology markets however, the volume extremes between a large buyer and the rest of the market could be vital. Volume purchasers often have a different set of requirements to regular buyers, not just in terms of product features, but also in management of those products – how are they to be handled (shipping, installation, delivery timing etc), how is invoicing allocated, what volume discounts are available, how are training and installation to be dealt with, and how is service and support to be handled. In technology markets, frequent buyers are more likely to be early adopters and so the features they look for might guide the rest of the market.

So in these circumstances stratifying the sample by volume to ensure adequate data about volume or frequent purchasers is absolutely essential. And in analysis this volume distinction needs to be drawn out in detail as the top 20% of the market may be influencing far more than 80% of the sales. Strategies like 1 in N for this type of sampling, may be better dealt with by trying to contact all major customers, setting incentives or involving key account managers in helping to get the research done. For instance, a small business buying mobile phones may have a focus on make, model and broad operating costs. A large company on the other hand will have issues with asset tracking, billing (and so allocation of expenditure), require a broad geographic service network and look at factors like technology integration with existing systems. If you have 80% small businesses and 20% large businesses in your sample, the lesser needs of the small businesses may lead to focusing on elements that are not so important to the volume-part of the market.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: