Crowdsourcing a Forecast? Ran a Twitter contest about Microsoft in Smartphones

I had an idea on Twitter last week and thought it would be a fun little game among some of my most dedicated followers and a way to give away a nice little freebie gift. Since I have been quite public about how I think the Nokia and Microsoft partnership will fare in the coming early years (in other words, I have been extremely pessimistic about it all) and I have said in public I expect Nokia to have about 8% market share in smartphones end of 2012 when all their smartphones should be running the Microsoft Windows Phone based operating system, and that Microsoft overall would only have about 10% at that time, I wondered what would be the guesstimates of my followers on Twitter.

So I asked my followers to submit their guesses. Almost immediately the response was better than I expected, to I refined the contest, to give the guess to the accuracy of one tenth of one percent (ie 8.4% or 12.3% etc). Also as I would be giving away a prize (free ebook) to the person who guessed the closest, I also said that any one number % guess, would be given on a first-come, first-served basis, so whoever was first to claim a given exact percentage, would have that number and be the only one obviously eligible to win the prize at that specific number. Any Twitter follower would be allowed only one guess. And as some started to request numbers that were already taken, I would then tell them which were the nearest numbers above and below their request, that were still available.

This was all fine and good until about 120 entries had come in over the weekend and I found that several whole ranges of numbers were taken. First to go were 7.x and 8.x, soon also 5.x, 6.x and 9.x went. By Tuesday morning we had over 150 entries and all numbers from 4.6% to 13.4% were taken and 14.x also was all gone.

I had also thought that this would be an interesting test of crowd-sourcing, the wisdom of the crowd, Alan Moore's 'Nobody is as clever as everybody' etc. But I had unwittingly spoiled part of the precision of this 'instrument' in forcing people to pick unique numbers. Why didn't I let people pick multiple times the same number if they wanted, I could have easily given them each the winning prize as its an ebook haha, no real printing costs for me, etc.

I'm sorry I was not thinking. This could have been far more revealing if we had been able to see where the guesses were clustered, if many for example picked 8.4% and 8.6 % or whatever..  And by Tuesday, so many of the numbers were gone, that some who wanted a number, were now forced many whole numbers away to the 'nearest available' number.

What I did do, was to back on Twitter to see all submissions, to try to collect the 'first guesses' ignoring if those were taken, but unfortunately there were so many comments that I couldn't get back to the beginning. That would have been probably more telling than just this list we have now.

So, first, for all who have already entered (or who may still want to enter) here is the official entry list of all who had submitted a guess and gotten a number by Tuesday 8 November 2011. What is the market share in smartphones of all Microsoft based operating systems sold in Q4 of 2012, including Nokia phones running Microsoft (but ignoring all non-Microsoft Nokia phones such as running Symbian, MeeGo etc) and any other smartphones running Windows but ignoring tablets and PCs. Each entrant is up to accuracy of one tenth of one percent. The names here are the Twitter handle and in parenthesis the name on TW behind that handle. The official entries are:

Tweetvote of what will be Microsoft Windows Phone 7 market share at end of Q4 of 2012, a little over a year from now:

0.4% - @3GDoctor (David Doherty)

0.5% - @MyInnerSperse (Bruce Burke)

0.9% - @tonyfish (Tony Fish)

1.0% - @BenFletch (Ben Fletcher)

1.1% - @Dutchcowboy (M Lens-FitzGerald)

1.4% - @gibtang (Gibson Tang)

1.6% - @josiefraser (Josie Fraser)

1.7% - @miusuario (Omar Moya)

1.8% - @r_g (Ged Carroll)

1.9% - @lukabirsa (Luka Birsa)

2.0% - @rionov (Rio Novrianto)

2.1% - @jmacdonald (Jonathan MacDonald)

2.6% - @simplynand (simplynand) - said 2.63%

2.7% - @eldarmurtazin (Eldar Murtazin)

2.9% - @Starlo777 (Steve D)

3.1% - @dougsymon (Douglas Symon)

3.2% - @devonseaglass (Ian James Cox)

3.3% - @edent (Terence Eden)

3.4% - @jakublipinski (Jakub Lipinksi)

3.5% - @virtopeanu (Dan Virtopeanu)

3.6% - @ClaireBoo (Claire Boonstra)

3.7% - @cibyr (Ian Cullinan)

3.8% - @mobilezeitgeist (Heike Sholz)

3.9% - @AgileArtem (AgileArtem)

4.1% - @sdemidov (Sergey Demidov)

4.2% - @ilocco (ilicco elia)

4.3% - @c1rrus (James Nash)

4.4% - @Pierkka (Pirkka Poole)

4.5% - @mobilesguruji (Ankush Kala)

4.6% - @juhariis (Juha Riissanen)

4.7% - @yeswap (Dennis Bournique)

4.8% - @Guamguy (Guamguy)

4.9% - @InvulgoSoft (Thomas Andersen)

5.0% - @kaimac (kai macmahon)

5.1% - @sirraH77 (Harri Viitamaki)

5.2% - @jamescameron (jamescameron)

5.3% - @sbepstein (steve epstein) - said 5.25%

5.5% - @hockamania (Daryl Goh)

5.6% - @Miss_mobile_web (anne thomas)

5.7% - @ian_martin (Pablo Medeiros)

5.8% - @NOMOSDesign (NOMOS Design)

5.9% - @tom866mot (Tomislav Tomic)

6.0% - @Madijn (Michael N)

6.1% - @HarveiBeili (Harvey Beilinsohn)

6.2% - @russellbuckley (Russell Buckley)

6.3% - @kapten_grogg (Anders Mellqvist)

6.4% - @michielb (Michiel Berger)

6.5% - @DHDe22ers (Dick)

6.6% - @PinnacleSalesUK (Pinnacle Software) - Said 6.55

6.7% - @ei_HMcCarthy (Hugh McCarthy)

6.8% - @leeomar (Lee Omar)

6.9% - @ju_key (Uwe Kaminski)

7.0% - @arnozinho (arnozinho)

7.1% - @r0gu (Rogu)

7.2% - @CWmobil (Computerwoche Mobil)

7.3% - @OWStarr (Oliver Starr)

7.4% - @bookmeister (bookmeister)

7.5% - @fribeiro1 (Fernando Ribeiro)

7.6% - @drostyboy (steven)

7.7% - @rabschi (Cornelius Rabsch)

7.8% - @Rock35 (Rock35)

7.9% - @nokleby (ChristiAN)

8.0% - @slackerninja (rishi alwani)

8.1% - @suarezjulian (Julian Camilo Suarez)

8.2% - @yousifabdullah (Yousif Abdullah)

8.3% - @phil_linttell (phil Linttell)

8.4% - @reyes (Mauricio Reyes)

8.5% - @NeilMilliken (Neil Milliken)

8.6% - @Teemu (Teemu Kurppa)

8.7% - @anshelsag (Anshel Sag)

8.8% - @baranyurdagul (baran yurdagul) - said 8.75%

8.9% - @O_NunoPereira (Nuno Pereira)

9.0% - @3fs (ThirdFrameStudios)

9.1% - @bhuwan (bhuwan)

9.2% - @wpruvot (Wandrille Pruvot)

9.3% - @ThatChrisSmith (Chris Smith)

9.4% - @kojacker (kojacker)

9.5% - @cybronics (Jerome Derozard)

9.6% - @indigo102 (Martin Wilson)

9.7% - @_sommer (Hrvoje Sommer)

9.8% - @jainrounak (Rounak Jain)

9.9% - @tomdunbar (Tom Dunbar)

10.0% - @mbrit (Matt Baxter-Reynolds)

10.1% - @Eshanse (Esben Hansen)

10.2% - @mattjbush (matthew bush)

10.3% - @ccoc (Colin Crawford)

10.4% - @khouryrt (Rita El Khoury)

10.5% - @karipelaaja (Harri Manninen)

10.6% - @barbaraballard (Barbara Ballard)

10.7% - @Pishba (Patty Hartwell)

10.8% - @eccella (Nigel Alexander)

10.9% - @AndyTow (Andy Tow)

11.0% - @rodrigottr (Rodrigo Arantes)

11.1% - @fneuf (fneuf gazouille)

11.2% - @aktiwary (Ashutosh Tiwary)

11.3% - @punkedge (Pankaj Bengani)

11.4% - @_wideangle (Juha Meronen)

11.5% - @wishblizz (wishblizz)

11.6% - @bartfuzzle (Bart Fussel)

11.7% - @garelaos (Gareth Jones)

11.8% - @pannkaz (Pankaj Kapoor)

11.9% - @raimundo (raimundo vazquez)

12.0% - @syamant (syamant sandhir)

12.1% - @seetu (Sitaram Shastri) also said 30M is number

12.2% - @billboxer (william fasse)

12.3% - @Mawkins (Mark Hawkins)

12.4% - @bernardosanchez (bernardosanchez)

12.5% - @janole (Jan Ole Suhr)

12.6% - @gcameron00 (Graham Cameron)

12.7% - @Markies (Mark de Kock)

12.8% - @cybette (Carol Chen)

12.9% - @teemuo (Teemu Ollilainen)

13.0% - @lesanto (Glenn Le Santo)

13.1% - @asrialbaker (asrialbaker)

13.3% - @bmkatz (Brian Katz)

13.4% - @mjjagpal (mj)

13.6% - @AlexeiBoiko (Alexei Boiko)

13.7% - @Haaza (Marko Haaja)

13.8% - @gcsanda (Gergely Csanda)

13.9% - @anti_nomy (Mike)

14.0% - @abouillot (Alexandre Bouillot)

14.1% - @darlingbuds (Andrew Darling)

14.2% - @lassi (Lassi Kurkijarvi)

14.3% - @kmtee (KM Tee) said 14.32%

14.4% - @MBBdV (muriel devillers)

14.5% - @tim_harrap (Tim Harrap)

14.6% - @talakem (talakem)

14.7% - @kimmolehto (Kimmo Lehto)

14.8% - @IzanCoomonte (Izan Coomonte Suarez)

14.9% - @markjenniskens (Mark Jenniskens)

15.0% - @slackerninja (rishi alwani)

15.1% - @RamziD (RamziDziri)

15.3% - @sonicnights (Neil Ramsay)

15.4% - @kaienmy (kaien)

16.1% - @thompabompa (Thomas Strindberg)

16.2% - @chip_mk (Petar Chichev)

16.4% - @spurcola (philip bland)

17.0% - @Rhymo (Raimo van der Klein)

17.1% - @adrahon (Alex Drahon)

17.5% - @zahidtg (Zahid Ghadialy)

17.6% - @mauricioiwata (Mauricio Iwata)

18.5% - @laowaijoha (joha)

18.6% - @verydeepsleep (Wyn)

18.7% - @msantoni82 (MS)

19.4% - @simon_staffans (Simon Staffans)

19.6% - @theovalich (Theo Valich) - said 15.1% WP7 and 4.5% WP8

19.8% - @Hari0o7 (Hari Gupta)

19.9% - @hephail (Werner Egipsy Souza)

20.0% - @autiomaa (Daniel Schildt)

21.6% - @RussB (Russell Beattie)

25.0% - @vincente @Vincent everts)

25.1% - @aape (Aape Pohjavirta)

25.2% - @teppohudson (Teppo Hudson)

27.7% - @barneyc (Barney Craggs)

28.2% - @tkoola (Timo Koola)

29.3% - @tjytweets (TJ)

35.0% - @hardeep1singh (Hardeep Singh)

43.2% - @snagdragon (ihatesheeps)

There are also guesses entered by the following people, who had selected a number which was already taken (and had not yet sent their next requested number):

@albertcuesta (Albert Cuesta)

@Ew4n (Ewan MacLeod)

@KenLittle (Ken Little)

@miusario (Omar Moya)

@pvesterbacka (Peter Vesterbacka)

@stephen_oman (Stephen Oman)

@ZecaJacintho (Zeca Jacintho)

The prizes? The person who has the nearest number will get three free ebooks of their choice from my ebooks series as published by TomiAhonen Consulting, ie my Pearls ebooks and my statistics volumes the TomiAhonen Almanac and TomiAhonen Phone Book (and any other similar ebooks published by the time the contest results are finalized)

The two nearest guesses who just missed, will both also get one free ebook as the runners-up.

The actual formal 'official' number will be based on the following formula - the total smartphone unit sales of Q4 will be the average of the four analyst houses reporting regularly smartphones sales per quarter ie IDC, Gartner, Canalys and Strategy Analytics. The actual units sold/shipped (no distinction to be made) of Microsoft OS based smartphones will not be the number reported by anyone else except any of those four analyst houses so any mysterious measurements by for example Microsoft or Nokia (like Google announcing its 'activation' numbers) will not bring any subject of controversy. Not all of the 4 analyst houses report OS market shares, but we'll take Microsoft market share as the reported unit number of those of the four analyst houses who actually do report it either in units or as market share percentage (calculating the unit sales out from that). Again, the average of those reported.

So it won't be 'Tomi's number' haha. it will be the mathematical average as reported by those big 4 analyst houses.

Now what 'did' we learn from the votes. The lowest vote is 0.4% and the highest is 43.2% but the average is only 10.1%. Remember the 'conventional wisdom' by many analysts and press seems to expect Microsoft in the 20% and above ranges. Out of all who guessed only 6% felt comfortable Microsoft would exceed 20% by the end of next year, so this very unscientific (and no doubt skewed) result seems to suggest a nasty 'realty check' to the Nokia and Microsoft watchers sometime after Christmas next year haha..

Obviously one has to point out that the sample is skewed in two ways - first, that most who read my Tweets tend to be rather seriously interested 'mobilistas' and very interested in mobile. The more the Nokia+Microsoft partnership results in pure 'mobile-like' and non-USA patterns of sales and adoption, the more this group might understand the likely future, but if the Microsoft partnership pushes Nokia products more to an 'PC-like' (Microsoft legacy) pattern and perhaps business users and USA success, then my followers will perhaps be biased against that kind of pattern.

Another obvious bias is me - my followers have heard my repeated vitriol against the Nokia Microsoft strategy and some may have had their minds 'poisoned' or at least influenced by me. A random sampling of mobile experts who do not follow me or read my blog would very likely score an higher expectation of Microsoft's chances, on average.

Yet it is very interesting. Over 150 guesses made and the vast majority expect Nokia and Microsoft combined to be somewhere in the 6% to 14% range of market share - well well below the total of what Nokia and Microsoft had independently just nine months ago, before Stephen Elop put Nokia in its death-spiral and abandoning most of its market share.

If you want to make your entry and join this fun game, send me a Twitter comment in the open (not a private Twitter message) ie start your Tweet with @tomiahonen and send in your number you'd like to take - please check in the above list that your intended number is not already taken. I will accept entries until the end of November 2011. Then we have to wait a bit over 13 months to find out who was most accurate in forecasting Microsoft's smartphone market share for Q4 in 2012.

 

Comments

[@rodrigottr]

Tomi

Of course I believe the sample is biased. But we are talking about a ecosystem who had 2% in past year and competing head to head with other two or three ecosystems with much much more adherence then Windows Phone has today.

My question is: what does the windows phone can offer that none other competidor can offer now?

XBox?
Office Suite?
Nokia Maps?
UX?
The excelence of nokia hardware?

In most of these items both iOS and Android can offer something comparable.

AND

How can Windows Phone compete with android for the preference of those who has the biggest impact over the client that are carriers?

How can Windows Phone compete with Android for the preference of Asian OEMs? If they don't offer software differentiation and even charge the OS while Android is free?

How can Windows Phone compete with the stronger Apple reputation and own branded stores?

In my opinion Windows Phone has just a different UI with basically more of the same possibilities and a weaker ecosystem and weaker support of those who are closer to the consumer and have bigger influence over it.

Not enought power for turning the irrelevant into relevant.

Probably the best they will do is to use the rest of Nokia's sales force and reputation for selling well on europe, since the other markets where Nokia was strong, China, India, Brazil mostly can't pay for the price of the WP smartphones.

In the end perhaps I even was too optimistic with my vote of 11,0%

hahahaha

[eduardo]

I thought Windows Phone 7 might have a good chance against Android because the latter had a so-so ui. But ice cream sandwich is a vast improvement, so I don't see how wp 7 is going to have much success.

[incognito]

Since my company is involved heavily in mobile solutions development, for three years in a row we have an inside poll (employee sourcing? :D) what to expect from mobile platforms in the next year, so that the management can have an input from the inside as well. We're a company of shy-bit above 100 people and everybody is casting a prediction no matter of position or technical expertise. Now, given that most of us closely follow what's happening on the mobile market, and some of us certainly have our own preferences, our 'crowd sourcing' is also biased, but we had some pretty fair guestimates.

We're doing the polling at the end of Q4 for our company (June) so the predictions were for the next Q4. This is an aggregated prediction list (total market share) world-wide, not new activated devices or trends. It also doesn't consider versions of operating systems, just their types. We have a separate poll for all those but I don't have the results at hand, so here's the aggregated list:

June 2010: Symbian 31.2%, iOS 30.8%, BlackBerry OS 15.3%, Android 7.8%, Windows Mobile 6.2%, webOS 2.4% (unallocated 6.3%)
June 2011: Symbian 29.6%, Android 19.9%, iOS 18.7%, BlackBerry OS 13.1%, MeeGo 6.2%, Windows Mobile 3.3%, webOS 3%, Windows Phone 2.2% (unallocated 4%)
June 2012: Android 30.2%, Symbian 23.3%, iOS 14.1%, Windows Phone 9.2%, BlackBerry OS 8.9%, webOS 3.6%, MeeGo 3.1%, Windows Mobile 2.1% (unallocated 5.5%)

Of course, some of those are not viable anymore (or were not viable in the time the prediction was made for), but it gives some insight. As for the end of 2012, I personally don't think that WP7 will hit the 10% mark, unless Microsoft completely rework it, give more freedom to the OEMs for customization and value-added things and releases Win 8 that will extremely tightly integrate with WP, while providing the similar options for at least users with Macs. Chances of that happening are not exactly stellar.

I think Symbian will sink under the 20% mark (although Belle looks promising), and Android might rise to a whooping 35%. Grim fate awaits BlackBerry as well, although it remains to be seen what will they do with their new BBX-OS. MeeGo (well Maemo 'branch') might actually grab a couple of percents given how well the N9 is received, provided that Nokia stops with artificially created lack of supply and purposed stifling of it in favor of their Lumia line, as they are doing now. webOS, as well as Windows Mobile will be practically extinct by the end of 2012.

As for Nokia, they really are in a deep doo-doo and I don't think that Lumia line will save them - I'm willing to bet that they'll sell more Symbian phones than WP7 ones by the end of 2012, and not only that - they will have bigger profit margins on Symbian devices and the N9 than on their Lumia line. Their featurephone business will keep them afloat, but the question is for how long...

While I don't agree with some of Tomi's rabid fanboyism and some implied conspiracy theories, and especially not with the 'carriers hate Skype' motion (after all, Maemo/MeeGo had, and has the best Skype integration out of all devices), I fully agree that going with the WP7, especially the way Elop handled it, will cost Nokia dearly. They actually had much better chances with their 'burning platforms' than with the paper ship they've jumped on. If they did it quietly while forcing their current one-Qt-to-rule-them-all strategy, they'd at least have the time to gracefully jump on the WP7 bandwagon while making profits and having more leverage in negotiations with Microsoft, and having a viable backup plan if WP7 flops. They could even force Microsoft to accept Qt as a first-class citizen on their platform thus creating a graceful transition for their existing user and developer base. Of course, the question is would the WP7 even survive by now if Nokia didn't make a commitment to it back in February, but unlike Nokia, Microsoft has money to burn for risky moves - Nokia forgot that this is not the 2007 and that they really, really cannot afford `betting it all on black`.

The way it is now, Nokia is on the verge of leaving the smartphone (and future tablet) race. I at least hope they're not that crazy to completely abandon Qt and that they at least keep Maemo/MeeGo as a backup, otherwise they really have set sail for epic fail even if Windows Phone eventually becomes the dubbed third ecosystem - by the time it might, it would be too late for Nokia.

As for my company, we've halted all of our Qt investment and are now primarily focusing on Android and iOS, while keeping an eye on the WP7. None of us (at least developers) would be especially pleased if we're to be forced to develop for the WP7 given that we've had some bad experience with .NET development (and especially so-called professionals in that field) and that we hardly fit into that mindset. We even received free Omnia 7's to investigate the platform - I keep mine in a drawer, that system is just not for me not even as a user. We were having high hopes for Qt, but Nokia dashed them quite expertly. 'Bravo' to them...

Cheers

[cheapdrdrebeatsen]

I thought Windows Phone 7 might have a good chance against Android noly in his ui.

[cheapdrdrebeatsfr]

I will test the windows 8 , I hope it can take me something special!

[cheapest beats by dre]

Superior quality with low price! Free shipping with fast delivery.Cheapest beats by dre headphones sale With FREE SHIPPING,sound wonderful high performance headphones,buy now!

[Beats by dre shop]

Superior quality with low price! Free shipping with fast delivery.Cheapest beats by dre headphones sale With FREE SHIPPING,sound wonderful high performance headphones,buy now!

[eduardo]

@Baron95

Ok, you persuaded me, at least some.

But on the patents thing I wouldn't be so sure. See the Barnes &Nobel lawsuit, good coverage at Groklaw, Microsoft's claims look very weak.

[Mobile Prices in]

In my opinion twitter is a very good platform for Social Media and it is really workable and profitable for all whom are working on it.

[dissertation writing uk]

hm, i dont` know, but i don`t like Windows Phone.

[Nokia Mobiles]

Really twitter is a very big and great platform of social media and it helps to promote different types of news and websites all over the world.

[ugg]

This type of subject is very good for men and women to find out much more about it, and that men and women ought to be every single day much less ignorant, and medicine for this really is the reading of troubles like this.http://www.uggbootssalemall.com Thanks for you sharing..,

[prada]

Pretty good post. Ho appena imbattuto il tuo blog e volevo dire che ho apprezzato molto leggere il tuo post sul blog. Alcun modo sarò sottoscrivendo il feed e spero di postare di nuovo presto.

[Lancel]

Bonne chance à vous dans le futur. Vous m'avez aidé aujourd'hui.

[Hogan Scarpe]

Standard Poor's plans to update Hogan Scarpe its credit ratings for the world's 30 biggest banks within three weeks and may well mete out a few downgrades in the process, possibly surprising battered global bond markets.

[マジコン]

Your site is amazing.I am very impressed to see this,i want to come back for visiting your site.Keep doing Good as well as you can..

[Mobile Phone Prices in Pakistan]

Such an awesome post and it really increases my knowledge to high extend.

[Abercrombie]

Azioni di Abercrombie & Fitch è sceso 1, 15 dollari, a some, a couple of every cento, every chiudere any money 26. 10.Inoltre il Venerdì, JC Penney, il valore-prezzo negozio di reparto, haya detto che il profitto every my partner and i tre mesi che si è concluso a couple of maggio cadde il 79 every cento, any money twenty-five milioni, a 11 centesimi every azione, rispetto ai money 120 milioni, a 54 centesimi every azione, every il periodo di not anno fa.

[Woolirch]

Giubbotti Woolrich marrone every giovani adolescenti elizabeth gli anziani，Le giacche sono sempre my partner and i vestiti soprattutto realizzare gli esseri umani mite inside inverno, anomalo il Woolrich Doudoune. ELIZABETH 'stata l . a . tanto acclamata del forged every realizzare voi, è l'attualizzazione elizabeth reale elizabeth giubbotti Woolrich elegant. These kinds of consiglio si accumulano mite nel più duro del fermo invernale elizabeth già si abrasione uno giubbotti Woolrich vendita dei prodotti, si è apprendista presso not angolo parte della folla.

[Ken Griffey Jr Shoes]

Everything must be done to set a deadline.