What do we actually know about WotC's market research?

Minis and 3.5

Looking at the market research, I note that those who use minis spend a lot more than those who do not, and there was a significant chunk of the gaming population not using minis. Does this relate to the mini-focused rules in 3.5?
 

log in or register to remove this ad

Raven Crowking said:
Is this accurate? 1,000 respondents is far from exhaustive research, IMHO.

IIRC, that kind of thing is pretty standard. If you can generate a representative sample, then you test with a far more management group size while still maintaining accuracy (of course, some margin for error is introduced).

I'm not really a statistics expert though.
 

Raven Crowking said:
Is this accurate? 1,000 respondents is far from exhaustive research, IMHO.

Statisticians would disagree with you. The way the math works out, a sample of 400 gives you 95% accuracy (which is usually used in surveys). A sample of 1,000 gives you 97% accuracy in your conclusions. This is regardless of overall population size.

See here: http://en.wikipedia.org/wiki/Sample_size
 

Raven Crowking said:
Looking at the market research, I note that those who use minis spend a lot more than those who do not, and there was a significant chunk of the gaming population not using minis. Does this relate to the mini-focused rules in 3.5?

Very early in the process the idea of creating a gaming accessory that folks might pick up on every trip to the gaming store was discussed. Minis was a natural fit for this. Picking up some minis for the next game was akin to purchasing a MTG booster or a new set of dice. Obviously, the idea has changed some with the introduction of DDMs, but the recurring purchase part still fits.

In other words, I think 3.0 was pretty mini-focused and 3.5 just went a bit further down this road.
 

Hi, all--

The 1999 study is well known mostly because its results were shared, in part, with the public, and also because it was the first act of real, professional market research ever carried out for D&D. But it's a drop in the bucket for WotC, which has, since then (at least through a year ago, when I was running D&D), continuously conducted formal and informal market research to best understand what gamers want out of D&D.

Among other things, WotC conducts (or, again, did while I was there--and probably still does) an annual awareness and usage study, an ongoing consumer panel, focus groups for key products (generally things in which the learning process is key, such as the Basic Game or minis starter set), and surveys and interviews at big shows such as Gen Con. There's also market research done on the retail side of things, including a monthly study of retailer purchases and activities across the entire hobby gaming category.

WotC has professional market researchers on staff, and employs outside agencies. Regardless of our speculation about the validity of sample sizes, the techniques used are up to industry standards for this type of research.

Market research is just a tool, of course. Its results can be misinterpreted, ignored, or poorly acted upon. Even under the best of circumstances, market research is just one factor contributing to a good business or creative decision. The people who run D&D may or may not make great decisions, but either way they have a substantial amount of continuously-updated market research to call upon.
 

Raven Crowking said:
Is this accurate? 1,000 respondents is far from exhaustive research, IMHO.

Many, much more 'important', surveys are done with similar or even smaller numbers. For example television ratings are in my country of 5 million done with couple of thousand machines, and heckuva lot more money depends on that than D&D.

EDIT: Just checked - there are exactly 1000 of these ratings machines in finnish households. Advertisement prices are dependent on these numbers.

which seems to indicate that every postcard was intended to glean information on 3 or more individuals. In other words, for every "primary" respondant, information was given second-hand on 2-3 others, seemingly skewing the results drastically toward second-hand information.

Answering for same household is not really second-hand information.

Is there something here I'm not seeing?

Yes.
 

Numion said:
Many, much more 'important', surveys are done with similar or even smaller numbers. For example television ratings are in my country of 5 million done with couple of thousand machines, and heckuva lot more money depends on that than D&D.

EDIT: Just checked - there are exactly 1000 of these ratings machines in finnish households. Advertisement prices are dependent on these numbers.

Interesting. And I followed the wiki links that Delta supplied. Obviously, a sampling of 1,000 is considered ample to generate a result within a close degree of varience, and in many cases the statistical sampling should yield accurate results.

I've heard from time to time both that the baseline setup in 3e was based upon the statistics gleaned (evidence for which does not appear in the published reports) and that it was determined as a consequence of how 3e playtested. Anyone know which of these conflicting reports is accurate?


RC


P.S.: On the subject of minis, the WotC stats suggest that, if minis are used, there is a very significant increase in profit. The WotC stats also suggest that, using the wider spectrum of D&D editions in the survay, many people did not use minis. Finally, the stats suggest that there was very little overlap between the collectable mini gamer and D&D players. The creation of a set of minis/game that would appeal to two seperate gamer types was not only a no-brainer for WotC (even brought up in one of the reports), but so is increasing the mini-dependent nature of the game to boost sales.
 

Numion said:
Answering for same household is not really second-hand information.


So, you are basically saying that if I answer a survey about what my partner or my children like, the information is as accurate as if they answer? Or the other way around? :confused:

Any information that does not come from the primary source is second-hand by definition.


RC
 

I'm part of a few on-line survey groups (non-D&D-related at all) that sends e-mails to do surveys. They are commonly about food products, or other household items. Each survey always include questions regarding age/gender/race/etc, so that the company that runs the survey can classify responders and insure an overall correct proportion. Also, ceratin surveys target certain groups.

Anyway, about 6 months ago, I was extremely surprised to go to answer one such survey from one of the company I receive them and find out that it was about RPGs, and apparently (based on later questions) that WotC had ordered said survey.

It was mostly about awareness of different systems at first, then later asked specific questions about which books I owned, my buying habits, etc, etc.
 

Raven Crowking said:
I've heard from time to time both that the baseline setup in 3e was based upon the statistics gleaned (evidence for which does not appear in the published reports) and that it was determined as a consequence of how 3e playtested. Anyone know which of these conflicting reports is accurate?

They're both accurate, as far as I understand. The baseline setup was determined in accordance with the info gleaned, and then it was playtested, and the results of the playtest were no doubt measured against the yardstick of the survey data.
 

Remove ads

Top