Market analysis

[Razgon]

I'm interested in doing a market analysis of roleplayers, who, what, when where why, and that sort of thing.

Since I'm fortunate enough to be in a company that has people all over the world, I have acces to a lot of different countries, at least specific areas of these countries.

Although I have some experience in this kind of work, it's not really marketing analysis's as such that is my area of expertise.

So, what I was wondering was, if some of you excellent designers, publishers, artists, writers and so forth would be interested in some sort of joint effort here, to look at market trends and find out, if there really IS a market for what we do, and what kind of audience we really have and any trends they follow.

Let me know what you people think, ok?

[Patrick Chipman]

I posited a more scientific research-oriented version of this a while ago. The idea was to use a number of sampling methodologies to get a clear picture on what gamers are like, as a potential basis for future research. It would certainly go a long way to settle, once and for all, the burning questions in gaming ("Do those con people actually play real games?" "Do stealth gamers exist?" And, of course, the perennial question, "Are we better than normal people?" ;) ).

The problem is, most people consider this market research, rather than exploratory pure research. My intention never was and still isn't to collect data to build better-selling games; that's a very bad way to construct what amounts to art. It's also worth noting that the trend of the industry appears to be "buy more D&D," since it is the de facto standard. Is it possible to figure out why D&D is so popular? Sure, but it's not market research that's going to tell you that.

In any event, if you're interested in taking a scientific, non-marketing approach to this, I'd be willing to lend a hand.

[Misguided]

This is a splendid idea, if it can be pulled off. Research like this has to be carefully planned and constructed from the beginning, though, or the data collected could be worthless. You need experts on collecting this kind of data to do it properly.

[Razgon]

The company I work for, is a marketing company, so I do have some experience in this kind of work, although not much hands on unfortunately.

I'm sure that with a few interested people who knows a bit about the industri, we can pull this off.

I will volunteer as being project manager, unless someone more qualifed come along, since this is what I do for a living :)

[Wingnut]

This is a great idea in theory, and something part of my business plan for Impressions for years. The only problem is the execution of something like this into the marketplace.

The methodology to do this is broken unless you survey gaming in general and NOT specifically, paper vs. electronic games. If you do paper games specifically with the general marketplace, you'll get Monopoly and Trivial Pursuit answers.

WOTC did a survey, but its biased towards their consumers.

You cannot do an online survey, then its biased against non-computer users.

You cannot do a direct mail survey unless you take a random sample of consumers from across the U.S....and forget about getting them from retailers then your methodology is biased towards retail shoppers.

It's a problem. I even tried to do some type of survey on my GamePlay Demo CD to 10,000 people and the response rate was too low. Imagine how hard it will be to try and find 25,000+ generic consumers not biased towards anything...

I hope that helps.

Best,
Aldo Ghiozzi
Owner
Impressions Advertising & Marketing
www.impressionsadv.net
Wingnut Games
www.wingnutgames.com
Game Publishers Association - Director of Marketing
www.thegpa.org

[Razgon]

Well, one way is to hand out questionaries in hobby shops all over the world :) Or at least a representative portion of many countries.
This could be done actually. We here at Lidium live in 5 different countries, so we alone have acces to a large audience, if other people join in, we can do a lot here

[Misguided]

Any sample will be biased in some way, most likely. Bias isn't evil. Not KNOWING what the biases are is bad. Trying to draw conclusions on a population that differs from the sample can be bad. Personally, I wouldn't mind a sample biased towards retail shoppers, particularly if the stores were chosen with some care. I would love some well collected info about the habits of shoppers at FLGSs.

[Patrick Chipman]

First, le tme say that I'm finishing up my B.Sc. in Psychology, with a focus on experimental methodology, and I've run and designed a number of experiments in the past. So I know something of research. ;)

Anyway, here's the methodological problem I see with this: a self-report, self-return survey is going to have low response rates and will be biased to those who want to respond. Being nonconfrontational and with no apparent reward, respondents will be those who want to "make their views known." This equates to a 10% return rate, if that. Additionally, self-reports are notorious for being inaccurate, whether it's through lying (for any number of reasons) or simply not knowing one's own behavior.

[Razgon]

Doesnt have to be like that. I imagine if we are a a few of the people from in here, deciding to do this, we can offer a few prizes as well.
Shouldnt be a problem.

I know that the best thing would be, to be there oneself, and ask the questions, but since this isn't practical, what we can't get on quality, we can get on quantity. The more answers, the better a picture we will get

[Patrick Chipman]

Adding prizes may improve response rate, but it isn't going to improve the quality of the responses. All the data in the world isn't useful if it isn't valid. ;) (Well, that's not entirely true; more data points equals more power, certainly, but a better way to get power is to increase the effect size with a more sensitive test.)

That having been said, I think one could get volunteers to construct a large, non-profit database using multiple methodologies. When I say "confrontational," I don't necessarily mean being physically present; you can cold call (which is the methodology used by polling agencies), passively observe (useful for determining the behaviors of game store patrons), or use other techniques.

Generally speaking, survey research isn't as valid as other methodologies. It's useful, but only really as a starting point; other techniques need to be used to "zoom in" and get more valid and specific data.

[Misguided]

The real question is what would the purpose of such an effort be? What do you want to use it for? What kind of conclusions do you want to be able to draw from it? That will guide the design process.

[Razgon]

Well, the purpose if the market analysis would of course be different aspects, according to the wishes of those participating in it.
I think we can all agree, that we are not going to make our "art" according to public opinion only, but I think we can all agree, that since most of us counts on getting some profit on this, it would be good to know where one's core audience is, age groups and so forth, in order to market the product accordingly

Besides that, I think it would be i interesting to see what trends actually are today, out of sheer curiosity of the business we are in.

[InvisibleJon]

A poor, Misguided individual wrote:
Bias isn't evil. Not KNOWING what the biases are is bad. Trying to draw conclusions on a population that differs from the sample can be bad.

Well-phrased, Lewis! This is such a critical point to understand.

(Freakin' USA Today surveys with a sample size of 607 people surveyed in downtown New York and turned into an infographic. Grrrr.)

[Misguided]

Originally posted by InvisibleJon


Well-phrased, Lewis! This is such a critical point to understand.

(Freakin' USA Today surveys with a sample size of 607 people surveyed in downtown New York and turned into an infographic. Grrrr.)

Thanks, Jon. I do have a LITTLE experience in this area... :)
I'm not sure there is such a thing as a truly non-biased sample. What you can have, is a sample for which it is valid to draw conclusion from given a particular target population. It all depends on how that population is defined and the sample collected. Don't mean to rant or get technical.

Again I return to the question of what questions do you want to ask and what population do you want to be able to make inferences about. This is where intelligent research design should begin, IMO.

Why isn't there more marketing data that pertains to this industry? Collecting useful data like this requires considerable expertise, time, and other resources.

[Razgon]

Misguided, and all the rest of you people who are interested in this...
The goal of the market analysis is to get a clear and concise overview of the market, to help newbies, old timers and everyone to better market and publish their games.

There are a lot of people responding to this thread who appearently has some experience or training in this sort of thing.

I propose we set up a forum (I can do so over at Lidium's (www.lidiumonline.com) forums), discuss this out, and give it a try, and make the first market analysis in this area by others than WoTC own internal marketing dept.

Let me know everyone what you think of this, and whether you want in or not.