Archive for the ‘Market Research’ Category

Who cut the cheese I mean market share?

April 29th, 2009 2 comments

Got a sinking feeling that something isn’t right? Something stinking up the joint? Well check your cupboards for some market share.

If you read my post about sizing up the mobile phone market, you may have gathered that I treat market share cautiously.  So the topic today is how to handle market share without soiling your hands, or your co-workers hands, or your boss, or loved ones.  Get out your vertical reference tools and read on.

Over the years we’ve all been bombarded by articles and reports and analyst commentary that said Microsoft’s IE browser has the most “market share”.  It’s so pervasive, I’m not even going to bother to site a link to a report.  It’s axiomatic, it’s common knowledge, it’s part of our heritage.  Keep that in mind for a second.

Today’s example of research comes from our good friends at TechCrunch.  MG Siegler wrote a piece today about a company called Wakoopa.  The gist of the story is that Wakoopa has a user base of 75K users, and they track how much time those users spend on activities and applications (according to the article the data is gathered from desktop tracking software users install).  I think it’s a great article, really interesting, and of course it provides a bit of grist for our mill here.  So after you’re done here, I recommend you read it.

Along with various other data cited in the article, the Wakoopa data shows that “Mozilla’s Firefox browser … is the number one app on both Windows and Mac platforms. And in every continent besides Africa, it’s over 60% usage among web browser“. The article goes on to say, “… Africa is the only continent where Internet Explorer finishes as the second most-used browser. On every other continent,  IE is in third place among browsers — or worse“.

Now wait just a minute!  That cannot be true.  We all know, with certainty, that IE has the most market share.  So (as Elaine would say), “get out of here!”.

Well, IE probably does have the most market share.  But the key question is, “share of what”?  And that one question is the most important question you can ask when presented with market share data.  Share of what? The TC piece on Wakoopa forces us to ask that question.  The “market share” IE has is really about installations on desktops, not about usage.  A very important distinction.

I cannot say this enough: “share of what”?  There are other questions, like for example is the Wakoopa data statistically relevant, what regions are covered (WW in this case), how does the desktop tracking actually work, what platforms does it sit on, how often is the data updated, and so on. And without reading the Wakoopa data in detail, and the methodology that drives it, you really don’t know what you have. So when dealing with market share, be careful to make sure you are standing upright.

Thanks for stopping by!  And remember to keep your vertical reference tools handy.

Categories: Market Research Tags:

Sizing Up the Mobile Phone Market

April 21st, 2009 2 comments

PlumbBob Research Sizes Up the Mobile Phone Market:
Androids, iPhones, Louis XIV, and the aesthetics of reason

A few weeks ago a friend of  mine who runs a software development company (Agorasys)  and I were talking about developing a mobile phone application (coming soon!) and we got to talking about which mobile phone platform would be the best one to develop on.  For a couple of reasons we settled on Android.  One reason was that we already had a Google phone, but that said we still struggled with the decision. We felt that it was important to get a sense of where the smartphone market was going, what the potential market was, how the application would be marketed, and what kind of application to develop. We assumed the market was dominated by a few players (i.e.,  Apple, Google, Nokia, Rim, Palm) and we understood each of these companies has strengths and weakness, and target audiences (i.e., consumers, enterprise, regions, etc.). So we set out to do some market sizing, forecasting, a little competitive scanning, and an assessment of how easy or hard it was to do business with the various companies.  What we wanted was a presentation with a few slides to help us focus.

So the specific topic for today is the smartphone market, and the role that market research plays in driving business decisions. In this case we have a microcosm of a very typical situation: a product development team needs to assess the market and see how to apply their core competencies to the current set of market dynamics; in other words the team needs to see how to make money doing what they do best.  We needed to determine what product consumers would pay for, how much they would pay, which platform to use, how to market the product, what resources would be needed (e.g., budget, time), how to structure a partnership, and determine what risks were acceptable and which one were not.  Once we did that, we would have to step back to the original question: what would lead us to believe we would be successful in this particular endeavor?

Before we start getting into the market research discussion, let’s do a warm up exercise, something to limber us up so we don’t pull a hamstring.  So read on.

Did you know that Louis XIV (“Louis the Fourteenth”) was a furniture maker? Probably not.  It’s not a well known fact.  But it’s true, and I’ll prove it to you.  First, do you agree that Louis XIV existed?  He is a well known historical figure, so there really isn’t much to dispute — he was King of France in the 17th century.  So we’re good there.  Next,  have you ever heard a piece of furniture referred to as “Louis XIV”?  Most likely you have, but if not it’s easy enough to verify —there are hundreds of thousands of Google hits, depending on what terms you use.

So there you have it — Louis XIV must have been a furniture maker!  How else would his name have been associated with all those chairs, tables, and so on, unless he were a furniture maker?  Not convinced?  Well let’s restructure the assertion and create an easy way to see the logic:

  1. It is an established fact that Louis XIV was King of France.
  2. It is an established fact that “Louis XIV furniture” is a well known style.
  3. Therefore Louis XIV was a furniture maker!

So now that you have the facts, you must agree with the conclusion, right?  No? Well then we’re at a dead end: I state some facts, you agree with them, I assert the obvious conclusion, and yet you don’t agree.

Stalemate? Well maybe not.  Instead of asserting that Louis XIV was indeed a furniture maker, what if we asked this question instead: what role, if any, did Louis XIV have in influencing the development or definition of Louis XIV style furniture?  As it so happens that is a good question for two reasons: first, it opens the conversation up and allows for new conclusions, and next because it helps us understand how our attitudes (i.e., our bias) and methods of reasoning shape our conclusions.

It turns out that there is some evidence that Louis XIV played a significant role in Louis XIV style furniture.  The Sun King, as he was known, wanted to establish a powerful political dynamic with himself as the center, and he used furniture to show how wealthy and powerful he was (his opinion was that kings had a divine right to rule):

“Louis XIV set up a furniture industry on the outskirts of Paris for the sole purpose of creating political dominance through artifact. Design motifs backed the king as all powerful, and furniture was interlaced “L,” fleur-de-lis, and the sunburst–Louis XIV was known as the Sun King. The French king was not only advertising his power over the church in these furniture designs, he was also positioning himself as a semi-deity to his people.” [Pocket CEU: Understanding Louis XIII, XIV, XV, and XVI Furniture, by John Kroger]

Now how do we know this is true?  In other words what credibility does the author (John Kroger) have? If you search Google for “louis xiv furniture” the page cited comes up 6th, with 5th being from the same domain.  Given that there are 361,000 hits for that search, being 5th and 6th is not bad.  But there is no Technorati Blog reaction, and not any useful Google backlinks either.  Apparently the article was originally published in “Fine Furnishings International Magazine“, and while there are several references to the magazine, FFI does not have much of a web presence itself.

That’s because FFI is no longer published.  I determined this by calling a phone number I found on an old web page (which at the request of the company I am not publishing); Mr. Kroger was gone, no contact information was to be given out, and no one would be able to help me with Louis XIV.  Dead end? Well not quite.

I found other references, from the Encyclopedia Britannica, and the Columbia Encyclopedia.  The Britannica says that:

At the Gobelins factory, founded by Louis for the production of meubles de luxe and furnishings for the royal palaces and the public buildings, a national decorative arts style evolved that soon spread its influence into neighbouring countries. [“Louis XIV style.” Encyclopædia Britannica. 2009. Encyclopædia Britannica Online. 20 Apr. 2009 ]

The Columbia says that:

In 1662, Louis XIV purchased the Gobelins manufactory and there Colbert united all the royal artisans, creating a royal tapestry and furniture works … [Manufacture nationale des Gobelins.” The Columbia Encyclopedia, Sixth Edition. 2008. 20 Apr. 2009 ]

Both sources confirm the original assertion of John Kroger that Louis was asserting his power as king; in other words his aesthetic dynamics were a means to a political end.

Given the facts at hand it would seem unlikely that Louis was an artisan, and while we cannot confirm his actual role in the design, it can be said with authority that he was the producer.  So there you have it: Louis XIV was not a passive bystander, but an active and aggressive participant in the style that bears his name.

The story of Louis XIV’s part in the furniture business illustrates a very important concept in decision making in general and market research in particular.  The way we go about structuring our questions, and our research methodology, has an effect on our conclusions.  In other words it is possible that the way we think about something determines our conclusions in spite of, not because of, the facts at hand.  So we need to be careful not to let our bias and un-articulated assumptions get in the way.  In the case of Louis XIV the original conclusion that he was a furniture maker (i.e., an artisan) may have been wrong, but it was just as wrong to dismiss the idea out of hand.  Turn the question around, and you get closer to the truth: Louis XIV greatly influenced the industry.  How we arrived at this new conclusion was to open our minds up enough to be able to ask the right question, and then find credible sources to get information.

So, onward to the business at hand: the mobile phone market, specifically the smartphone market.  It didn’t take me long to find information on the mobile phone market; in fact that was the easy part.  Making sense of it was the hard part.  The mobile phone market, including smartphones, is covered by IDC, Gartner, NPD, In-Stat, Juniper, ABI, iSupply, Forward Concepts, Informa, Strategy Analytics, Net Applications and probably others that I didn’t catch.

As we learned from the Louis XIV exercise, introducing bias, along with not acknowledging assumptions, can skew our conclusions.  When you are looking at research from several different sources, you have to keep in mind that each piece of research had a methodology, a set of assumptions about what market dynamics are in play, a set of definitions for product categories and market segments, assumptions about what will enable or disable the market, forecast models and statistical models that influence the final result.  Unless you know what questions to ask, and/or have a close relationship with the project managers and analysts who participated in the research, you are not going to understand much about the framework that led to the final product. You will not be able to simply cut and paste your way into a conclusion, because the individual pieces will not fit together. So be careful.

The big picture for mobile phones is that worldwide there were about 1.2 billion mobile phone shipments in 2008, according to Gartner.  This is roughly in line with IDC’s count of 1.18 billion units shippped.  So mobile phones are a big market when you roll up all the revenue streams (e.g., hardware and services). Smartphones account for 10-15% of that market depending on who you follow: Gartner says 139 billion smartphones, Informa says 162 million, Forward Concepts says about 145 million, iSupply says about 174 million.

But we still don’t have any context — do those numbers represent growth or decline?   Well, IDC says:

The worldwide mobile phone market experienced an unusual downturn in the normally robust fourth quarter of 2008 (4Q08) … the mobile phone market may not recover until later this year, and possibly not until 2010

But for smartphones: “If there was one highlight in 2008 it was that the converged mobile devices segment (commonly referred to as smartphones) grew 22.5% over 2007, clearly outpacing the rest of the industry.

NPD agrees that smartphones did well, capturing 23% of US handset sales in Q408, compared to 12% in Q407.  Notice that we have switched regions here: NPD data is USA, the other data points are worldwide.

Another IDC quote sums it up nicely:

…. In the years to come the industry will undoubtedly migrate more toward the converged device segment [note: this is IDC’s definition for smartphones]… The notion that this segment will remain in positive growth while the industry expects an 8.3% downturn speaks volumes about the potential upside for these devices when the market turns.

So the mobile phone market hit a bump, but smartphones are looking good.  The other research vendors I reviewed agree that smartphones will continue to grow and capture more of the mobile handset market (Juniper, In-Stat, iSuppli, Informa, Forward Concepts).

So, the smartphone market is looking good.  But it would be wise to ask a few questions about what we just read — two important questions would be what regions were covered, and how did each vendor define “smartphone”?  We should also ask did the sources get quoted correctly, or did anything seem out of context.   Determining whether data was worldwide or regional seems simple — do you think you won’t make that mistake? Be careful — Michael Arrington at TechCrunch did when he was discussing Hitwise data vs. Google data.

Understanding exactly how each research house defines “smartphone” would require careful reading of each research vendor’s methodology, but given that the references cited iPhone, Google phones, Blackberry etc. my feeling was that they were all talking about the same thing (but that said, see this comment about Gartner’s defintion from

Once you get past details about details, the next question is, what did we learn?  The simple takeaway is that the market for smartphones is growing as a percent of cell phones, and it’s in the hundreds of millions.  How can we apply it to our problem?  We could assume smartphones as a segment are here to stay.

What about Android as a development platform?  Well it looks pretty rosy.  Opensource platforms are projected to dominate the market according to Informa, with Android being the beneficiary (also from Informa)In-Stat says the same thing more or less:

“Growth in the smartphone market over the next five years will be led by Linux,  …’Strong demand is being driven by device manufacturers leveraging open OS devices to re-invent the mobile phone experience,’ stated Frank Dickson, VP, Mobile Internet Group at In-Stat.”

Now saying “open source” is not the same as saying “Android.”  In fact it’s not even clear what “open source” really means here (one would have to examine the documentation for Symbian, Linux, Andriod, LiMo Foundation and etc. to understand how each one approaches the concept).  But Linux and Android share a common DNA, so it would seem to be a positive development.

But what about use models, customer demographics, business models?  What about competitive forecasts?

According to NPD:

“As the AT&T 3G network construction continues, and T-Mobile’s begins, high-speed data is becoming more central to smartphones. In fact, two-thirds (66 percent) of smartphones now use 3G networks, compared to just 46 percent a year ago.” [note: remember NPD is USA centric]

According to ABI:

“70% of respondents used their mobile phones to access information through Internet search engines last year, reports ABI research from data in surveys in 2007 and 2008.  That represents a substantial 14% increase over 2007. The consumption of mobile data services, including search, has steadily increased year over year as indicated by the use of email and web from a mobile phone.

‘Use of the mobile phone goes well beyond voice calls,’ says [senior analyst Jeff ] Orr. ‘A significant percentage of respondents say they use their handset for mobile email and web access, reflecting the broader access to inexpensive or all-you-can-eat mobile data services in the US.’ ” [note the reference to US].

And from Informa:

“2013 almost 300 billion transactions, worth more than US$860 billion, will be conducted using a mobile phone – a twelve-fold increase in gross global transaction values in just five years.

Consumers are becoming more familiar with the use of the mobile phone for applications beyond calls and are more confident in the use of electronic commerce (via their experience of online shopping). Furthermore, there are signs that the key players in the ecosystem are now more prepared to collaborate and invest in creating the systems, infrastructure and consumer confidence necessary to ensure this market thrives.

According to [John Darnbrough, Associate], “in the developing world, the behavioral change amongst consumers has already begun; mobile payments and mobile banking are already the natural and only financial services to millions of previously unbanked consumers. “

Given the data we have so far one could infer here that the smartphone is growing as a segment of mobile handsets, consumers are using high-speed data services to access information more often then before, and the smartphone as transaction manager are increasing.  If that is too far a leap, then one could at least say that multiple sources indicate smartphone use is becoming more common.

So the market is large, smartphones represent a strong growth segment, and smartphone use models are widening.  I would also say that smartphones represent a decommoditization of the market, but that’s just my feeling.

Well then, let’s just go out and code the application and start making money.   But there are three questions still looming, and one is huge.  First, what about the various competitive threats — which platform is really best. Next, who is easiest to do business with? And the big question: who is going to buy our application anyway?

Who is going to win — what to do the competitive forecasts look like? After reading all the various news clips, references to reports, analyst opinions, and commentary, my conclusion is everybody and nobody.  I think the mobile handset market is growing, smartphones are growing as a segment, and the market is extremely competitive.  It’s anybody’s guess as to who will win; and I would regard any predictions about that as suspect.  If I had to guess, I would say that all the major players you see today will be here in 5 years, and we’ll see a few more in the game that are not there now.

By now you may have noticed I have not introduced “market share” as a metric.  First, I didn’t really think market share would drive our decision making.  Secondly, market share analysis is tricky because there are so many different factors to consider (e.g., what product segments to include, how to define product segments, regions, vendors, time frame, data collection methodology).  Constructing a useful market share model can be difficult, time consuming and full of bias, and we were looking for a quick turnaround. Should we look at hardware unit share, share of internet traffic, operating system share, demographic share, or what?  For our purposes I think the best share metrics were smartphone share of total mobile handset market, which I felt was adequately covered, and making sure Android was not going to fall off the face of the Earth.  In other words big picture stuff; and it appeared to me that all the vendors were pointing in the same direction, up and to the right.

But for those who are interested in vendor market share, here you go: share of US smartphone traffic.  Not surprisingly (a biased statement if ever there was one!) the iPhone is on top, RIM and Palm following, with Android coming on very strong.  One caveat: read the responses to the article before you jump to any conclusions.

Market share considerations aside, another big factor was the business models that are in place. For mobile phone developers, iPhone represents the most advanced business model as near as I can tell, but they exact a hefty price: 30% of the sale price.  But then again it’s the iPhone, with all the bling-factor and marketing.  Google, Blackberry and Nokia have appstores, less mature but they are there.  But in the short run it came down to Apple and Google.  My feeling was that Android would start popping up all over the place, and even though Google charges the same whopping 30% of the sale price, and you don’t get the bling-factor or marketing, we thought it would be a good bet.

And now we come to the most important question: why assume our application will succeed?  One answer would be that our decision making process was rigorous: we discussed several projects, we gauged the risks carefully, budgeted time and money carefully, analyzed our core talents honestly, and made our decision on which application to develop, what features to include, how best to market it and next steps.

But maybe the heartfelt, and totally biased answer is also relevant: because we want it to and we are willing to work hard to make it happen.  And if we don’t get rich this time then maybe next time.

Well that’s it for this post.  We didn’t cover every aspect of market research, not by a long shot.  But we have laid the foundations.

Thanks for taking time to stop by!   If you found any errors of logic, any links that didn’t work, or any failures of grammar or punctuation, let me know and I’ll make whatever corrections are needed ASAP.  If you happen  to have something to share, please do.

And remember to keep your vertical reference tools close at hand!