Raph’s Website » NPD market segmentation study

NPD market segmentation study

September 19th, 2006 (Visited 5468 times) Tags:

Another day, another market research report for all the marketing guys who hang out on this blog… This one is from NPD, and it purports to reveal that core gamers aren’t who the industry thinks.

Key stats:

45% of “heavy gamers” are between 6 and 17.
Almost 1/3 of “avid console gamers” are between 6 and 17 as well.
“Avid console gamers” was the largest segment.
They arrived at six segments:
- heavy gamers
- avid console gamers
- mass market gamers
- prefer portable gamers
- secondary gamers
- infrequent gamers
21% of “heavy gamers” are female
Heavy gamers make up only 3% of the gaming population
1/3 of the “prefer portable” group is between 18-34. But most of the rest are under 18.

Some thoughts… first of all, is anyone really surprised? Of course the base market for games is still kids. The much-vaunted “graying” of the gamner population is heavily driven by the increasing availability of casual games reaching other audiences, and not just aging demographic trends. But in addition, we mustn’t forget that 100% of young kids are gamers today. I don’t know a single grade-school aged kid who doesn’t play videogames.

Similarly, surely no one is surprised that in videogames as in all other industries, the hardcore make up a small percentage of the total?

If anything, the interesting factor is how much the “avid console gamer” is dictating the market, and not the hardcore. After all, the hardcore might be willing to check out Dwarf Fortress but avid console gamers certainly won’t — and not just because it’s a PC title.

The segments arrived at are interesting because they are not psychographic or demographic in nature at all, they are driven by actual play habits. Alas, NPD doesn’t seem to have anything from the report posted yet, so the above is all I have found. But I’d be very curious as to percentage breakdowns of these segments and demographics for each.

You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.

manicgiraffe.com selling pies and jam since 1986 wrote on April 9th, 2007 at 3:37 pm:

I was looking up prices for the Wii Nunchuk on Gamestracker and bounced off into gaming market studies by some people called NPD. You can read their bits if you want to, or you can follow me wandering after a long time onto this marketing blog, where a load of apparently important marketing people discuss it. I was actually trying to find if anyone had figured out how they defined their “six gamer segments: Heavy Gamers, Avid Console Gamers, Mass Market Gamers, Prefer Portable Gamers,

Faith wrote on September 19th, 2006 at 1:39 pm:

[...] Comments [...]

Raph’s Website » NPD offers a new gamer demo study wrote on July 5th, 2007 at 9:06 am:

[...] blogged about the original pass of this study. Now there’s an update. It’s focused on gamers specifically, and it looks like it was a [...]

Manhunt 2 Banned in UK - rllmukforum.com wrote on July 17th, 2007 at 5:53 am:

[...] be interesting to see the breakdown across all the groups and how much of a percentage they make up.This site has a bit more breakdown on the figures suggesting that the overall make up of the gamer [...]

Discovery Channel's Games Documentary Impresses wrote on November 27th, 2007 at 11:05 am:

[...] not sure if you’re lying or if you’re not remembering this correctly. Here’s a link to an article about the study. [raphkoster.com] To [...]

StGabe said on September 19th, 2006 at 10:19 am:

This doesn’t seem very useful to me without more information about their methods.

For example, how were gamers sampled and how did they avoid a bias towards younger or older gamers (this seems to be VERY hard to me)? What exactly does it mean to be a “heavy gamer” instead of an “avid console gamer” and how did they prevent bias there? I understand that they divided groups by analysing the clustering of data but that still doesn’t tell me much about how “hardcore” a heavy gamer is versus a secondary gamer or “avid console gamer”.

I went to the NPD website and couldn’t find the report. But that’s ok because it looks like I’d probably have to pay $5,000-10,000 for it even if I did find it.

Raph said on September 19th, 2006 at 11:17 am:

The article gives the following about their methods:

The study, whose sample is comprised of 16,670 participants between ages six and 44

The segments were determined by examining player behavior, rather than defining segments demographically.

10 behavioral attributes were used to determine the segments, and they fall into four primary areas: ownership of systems, use of systems, frequency of use, and purchase behavior. Segments emerged by minimizing differences within clusters and maximizing differences between the clusters.

StGabe said on September 19th, 2006 at 12:16 pm:

Yeah, I read all that. But where did they get their 16,670? Did they preselect certain ages or grab from one or another pool of consumers that was more likely to have certain ages (retail purchasers, for example)? And ok, they used 10 variables to determine segments, that makes sense — but where did each of the groups fall out? What was frequency of use, use of systems, etc., for the resulting segments?

Without knowing all that stuff it’s hard to understand what this study means. Given that they sell their studies for quite a bit of money, though, it’s not that surprising that the teaser article really doesn’t give us enough information to make solid judgements.

Allen said on September 19th, 2006 at 12:27 pm:

I know this is one of the few analysis services (I wont use data because thats hardly a valid data sample) available to the game industry, so I’m going to (begrudgingly) opt not to critically rip them to harshly, but I will say that I dont agree with their methodology or the price point for thier product. Thier (business) model is dated, hardly timely, possibly invalid and largely inaccessable to the industry. The PARK stuff seems more relavent although NPD did come to some decent conclusions (at least thier not extrapolating conclusions off a sample of 2000 static self selected survey recipients on this report).

Nex-Gen seems to flog NPD stats around fairly liberally (89 articles) so I’m sure they must have some traction in the game industry to some extent I just find a 10,000 price tag for 10 data points to be a bit steep (unless of course its a 100k person sample size)

Perhaps there are other meteors about to hit different sectors of the game industry….

mammals 4tw! ??

Darniaq said on September 19th, 2006 at 1:23 pm:

An important point, in my opinion, ois that this report, like so many, will be used by companies interested in getting or expanding into new spaces. Corporations aren’t known for their strict adherance to scientific validity Basically, the above allows people to spin what they want for their own agenda.

Is it factually correct? I’m not one to say. But it does corrolate with other stuff people have reported publicly and otherwise. And more importantly, it has the all important “common sense” factor. Common sense is not fact, but when fact can be used to support common sense, all the better.

A huge number of kids are playing games. This, to me, is because both a higher percentage of them are playing and, well, because there an even bigger generation.

Zaush said on September 19th, 2006 at 3:51 pm:

Simple equation (sort of)

Heavy gamer = gamer that plays alot = gamer that has alot of free time on their hands = kids and the unemployable

This is what I got out of that whole article. Not exactly “stop the presses” type stuff.

There is a diffrence between “heavy gamers” and “hardcore gamers” and it is not really about how much one plays games as much has what kind of games a person plays. The style of game is the contributing factor in my opinion.

And the fact console gamers out number PC gamers is nothing new. The system requirements for new PC games require a person to upgrade, or some cases, buy a completly new computer just to play. Not sure many famlies can afford a new machine every 2 years. Whereas a console has a much longer life span.

Having worked in the Market Research field for a number of years, I just don’t see much new information in this study.

Allen said on September 19th, 2006 at 4:37 pm:

Those are good points and your right Darniaq, Im just being a cranky old data/analyst troll. My main problem is the cost and accessability though for various reasons (Web 2.0, new business models, increased accessability to actionable data for decision makers as well as consumers). I didnt want to get into this but *sigh* here it goes:

10 data points x 10,000.00 = 1,000 dollars per data point, or look at it another way each person from thier self selected (static? paid?) survey pool cost the buyer of that report 1.67 cents
16,760 respondants represents approximately…
167,600 consumers, at a 90% accuracy
or 335,200 consumers at a 80% accuracy
or 502,800 consumers at 70% accuracy
or 670,400 consumers at 60% accuracy
or 838,000 consumers at 50% accuracy

How many video game consumers are in the united states? lets say 26 million
for ease of use? So therefore we can say conclusively that the NPD report in fact reports (10 data points) with accuracy on 0.032 of the video game consumers in the US at 50% accuracy. (Note: these numbers are admittedly rough statistical approximations, so I wont charge you 10k for them)

Well now if I can flip a coin and guess heads or tails on each data point about how the gaming market is segmented (less than 1/2 of 1% of the video game consumers I might add) tell me what exactly IS the value proposition of paying 10,000 for a report like this? You’d be better off trusting your gut instinct and experiance in this case.

Not to say analysis is always a bad thing even with small sample sizes, it gives you an idea about where things are “heading”, but it certainly dosent give one an idea of where things “are” right now. If you have large (and small) budgets and personnel resources to allocate for development 10k is 10k, it dosent matter if someone is the President of SOE or some Indie developer, they should get value for thier money, and so should gamers for that matter.

JuJutsu said on September 20th, 2006 at 6:06 am:

As another cranky old data analyst troll I have to dispute this. The reason your point about exactly who the respondents are and how they got there is so important is that it gets to the heart of the problem: whether or not they are representative of the population. With a decent random sample you don’t need anywhere near 16k respondents to get better than 90% accuracy for the entire population of the US not just a subgroup like gamers. Places like the Survey Research Center at the University of Michigan or the Dept. of Labor or the Census Bureau frequently do much smaller surveys.

On the other hand, if they have a really skewed sample I think your numbers are overly generous in terms of accuracy and credibilty.

While I’m ranting I may as well point out my reservations on the cluster analysis…I’ve seen over and over again cluster analyses that aren’t stable. Do the same clustering on a new data set and you get new results; run it on one year get 1 set of clusters, run it on next years data and you get a different set of clusters. Personally I want to see at least one ‘verification’ before I buy the results.

Whats the value proposition for a report like this? I dunno. I will say that this would never pass muster in the ivory tower.

Allen said on September 20th, 2006 at 7:41 am:

Well I revisited this on this fine AM because I realized my 0.032 or flippant “less than 1/2 of 1% of the video game consumers I might add” was quite off.

Anyhow Jujutsu, your correct small sample sizes can elicit accurate results…”On the other hand, if they have a really skewed sample” its my understanding thier samples are from a predesigneated “pool” of survey recipients.

I do tend toward preffering large data sets, particularly in an area where granular detail about the population is required (Healthcare), for the purposes of accuracy and detail. But every industry doesnt always have the data goodness of a 4m person database. It really isnt even the size of the datbase per se but the detail available in it that ensures robust and accurate analysis. So I didnt want to sound smug and come off as “yeah you need at least 2.6 mil unique records to be accurate” about the video game market because thats not correct, smaller “random” samples can be accurate as you noted.

“An unsophisticated forecaster uses statistics as a drunken man uses lamp-posts – for support rather than for illumination.”-Andrew Lang

Tholal said on September 20th, 2006 at 1:59 pm:

So which stats are bothering those who are complaining about their testing methodology?

JuJutsu said on September 20th, 2006 at 3:05 pm:

Qualm number 1. 2 studies, 2 methodologies, 2 different results. There seems to be some overlap, I assume that heavy gamers might equal power gamers and infrequent gamers might equal occasional gamers. Who has the better grasp of things? Dunno. I do know that I’m not going to put more credibility with NPD just because they 8 times as many respondents.
Parks Associates
Cai split gamers up into the following segments, based on time spent on games and motivation:
• Power gamers: This group represent 11 percent of the gamer market, but accounts for 30 cents of every dollar spent on retail and online games.
• Social gamers: This group enjoys gaming as a way to interact with friends.
• Leisure gamers: This group spends 58 hours per month playing games but mainly on casual titles. Nevertheless, they prefer challenging titles and show high interest in new gaming services.
• Dormant gamers: This group loves gaming, but spends little time because of family, work, or school. They like to play with friends and family and prefer complex and challenging games.
• Incidental gamers: This group lacks motivation, and plays games mainly out of boredom. However, they spend more than 20 hours a month playing online games.
• Occasional gamers: This group plays puzzle, word, and board games almost exclusively.
The NPD Group
The Video Gamer Segmentation Report separated gamers into six groups [“In all, 10 behavioral attributes were used to determine the segments, and they fall into four primary areas: ownership of systems, use of systems, frequency of use, and purchase behavior…”]:
• heavy gamers
• avid console gamers
• mass market gamers
• prefer portable gamers
• secondary gamers
• infrequent gamers.
Qualm number 2. Neither provides any info about how they got the respondents. With crappy samples, both could have different results that are equally meaningless.
Personal qualm. The NPD group didn’t have anyone over 44. I’m 57. What am I? chopped liver?

Aaron said on September 20th, 2006 at 5:33 pm:

Another relevant question is: Assuming these statistics are accurate, do those proportions describe only the present gaming population or untapped potential gamers as well?

I don’t think there’s an obvious answer. Baby Boomers, for instance, are relatively untapped (few, or no, non-arcade-style games designed for that audience) and a lucrative market (nearing retirement, with lots of money and free time). Culturally, they are significantly different from younger generations, so there’s a good chance that audience might break down into a different demographic pattern.

Allen said on September 21st, 2006 at 7:59 am:

Qualm 3. NPD wants to charge people 5-10k for a report the methodology of which they do not publish. Meaning you can spend 10k and might get your moneys worth or it might be 35 pages of superfluous inaccurate bad analysis. Worse yet is buying a report like this that ends up being full of terminology (analysis/marketing/etc.) but actually says nothing.

PS: Good breakdown Jujutsu

JD Deitch said on September 21st, 2006 at 8:46 am:

Hi everyone -

A couple of comments…

Allen - Your initial criticism does strike me as being a little bit… well… cranky, to use your own words. Good luck with your new business.

JuJutsu - Thank you for making some very good points (by the way, nice inferences considering you were working with just the press release). To your points…

Sampling (this should address StGabe’s question as well) – The survey was indeed conducted from our online panel and for this particular study we went out to 6-44 year olds. Demographically speaking the skews should be fairly well taken care of as we project and weight our data to be representative of the population, and we even go so far as to control for the interactions of various demographics that we find to be important to the space. You can find out more about it here (http://npd.com/about.consumerpanel.html).

NPD vs Parks Assoc - Our studies indeed look at two different groups of people, as you noted. Parks is looking at online PC Gamers. Our segmentation focuses on the universe of console + portable owners. Apples and oranges as it were. Having said this, we’ve also done research recently on online PC gamers and we have seen some similar things in terms of what appears to be an increasing group of casual gamers whose demographics are far more diverse than the typical ‘hard core gamer’ archetype.

Cluster analysis - You are entirely right here as well. Cluster analyses can be unstable, but there are ways of minimizing the instability. When I ran the analysis I did it multiple times with different cluster solutions and different orderings of cases. The solution we report ended up being quite stable in repeated tests, and we’ve actually applied it (via discriminant analysis) to datasets from two other surveys where we’ve asked the identical questions and gotten very solid results both for the new group we’ve applied it to and the control group.

A final note – Without tooting the corporate horn too much, we’ve been covering this particular industry for more than ten years, and the industry pretty much considers us the gold standard in terms of tracking and reporting both retail sales and consumer behavior. A lot of people spent a lot of time on this one. We stand behind the numbers.

- JD

JD Deitch said on September 21st, 2006 at 8:48 am:

Oh, and Allen, we do indeed publish the methodology in the report.

Dave Rickey said on September 21st, 2006 at 10:11 am:

NPD’s reports carry a high price tag, but once you step beyond the “true indy” level of financing, it’s not a lot of money. When you’re making multiple titles and considering what strategic positioning you should take, $10K isn’t a lot of money. If you’re a true indy making a single project because you believe in it you probably can’t afford it, but you don’t need it, either. Reading it probably wouldn’t change your mind about anything.

Whether they’re really worth it? Even bad numbers can be strategically meaningful, if your competitors are working off of them. If you’re a statistical wonk and you think you can do better for less money, have at it.

–Dave

JuJutsu said on September 21st, 2006 at 11:54 am:

Mr. Deitch

NPD expertise in sampling and weighting techniques ensures that return samples are representative and actionable.

That takes care of my sampling concerns. Your comments about the cluster analyses takes care of the stability concerns. Thanks for the reply.
Let me know if you need a 57 year old respondent….

Morgan Ramsay said on September 21st, 2006 at 12:34 pm:

Dave Rickey wrote:

If you’re a true indy making a single project because you believe in it you probably can’t afford it, but you don’t need it, either. … Whether they’re really worth it?

I don’t think market research isn’t necessary to any business; although, understanding the state of the market is useful to strategy. There is a problem with market research that isn’t a problem with the research. When businesses become reliant on the research, they tend to respond to market demands, creating a stagnant culture of incrementally innovative competition that is often referred to as being market-driven. Examples of these incrementalist offerings include Apple’s iPod, Microsoft’s Zune, Nintendo’s Wii, and Sony’s PlayStation Portable — all of which are convergent devices.

Firms that are market-driving are interested in offering radical innovations despite the risks inherent to straying from the status quo. The rewards of radical innovation are greater as we’ve seen with such businesses as Amazon.com and FedEx, but larger companies typically cannot afford the risks or the restructuring costs. Most market-driving businesses eventually become market-driven firms to retain their position in the market and to maintain stability, but some companies such as Sony (as a whole) remain consistent as radical innovators focused on growth.

Allen CFDT said on September 21st, 2006 at 1:16 pm:

Good stuff JD thanks for the clarity on various issues, missed the linky you provided r/t profile and population, so I’ll recant my population crituque. My cranky flippant data troll-self is mostly satisfied, I hereby apoligize for my uncaffinated early morning grumpy posts.

Dave-
Now Dave, I think you just issued me a challenge?

And I think I’ll take you up on it.

And here is my rebuttal challenge:
Please list for me all the stakeholders who need data related to video game players/market.

I have a certain predisposition for making things accessable to people who cant “afford” it. I like to think it’ll result in better games.

Signed.

Allen CFDT (aka Cranky Flippant Data Troll)

Web 2.0 Data Mammals 4tw!