Incomplete numbers

India’s premium smartphone segment grew 29% YoY in 2019; Apple was the fastest growing brand, up 41% YoY, and OnePlus maintained its No. 1 position

—The Economic Times (via Techmeme)

This headline is bothering me. It appears to present the idea that the Indian market is (finally) turning towards premium smartphones. But these numbers only present half the story.

It’s great the the premium smartphone segment grew 29%, and Apple’s sales (I presume) grew 41%. But from what base did they grow?

Also, how fast did the overall smartphone market grow? If the overall smartphone market grew by more than 29%, then the market share of premium segment actually shrank.

Half the numbers, half the story. Still makes it to the front page of Techmeme :/

Continue reading Incomplete numbers

Bias is good

Economist - Red, but not a red top
Red, but not a red top.

In professional cycling, the more dangerous surface to race on is not the slippery one, or a rough / cobbled one, but an unpredictable one. The cyclists can slow down, and/or change the angle of approach if they know about the slippery conditions, or broken road surface. Crashes happen when they go into a corner full gas not knowing about the bad road surface condition ahead.

Moving from pro-cycling to media (and people), the problems caused by bias are just the same. It’s easier to deal with a person who you know is strongly, and consistently biased one way or the other1 – irrespective of whether that bias is in alignment with, or against, your own biases views. It’s the unpredictable ones who are harder to read (and deal with). As interesting as people are to write about, in this post, I’ll stick to media.

Karthik wrote about what he considers The Economist’s *shit* coverage of Indian politics, tinted by their bias against Narendra Modi. Yet, he continues to read the newspaper2 for their non-Indian-politics coverage. However, in view of his recent discovery of the Murray Gell-mann Amnesia effect, Karthik ponders whether he can trust The Economist’s viewpoint on other topics anymore. Read his full post here.

The Predictable

My take on this issue is different. It relies on known and unknown biases. Having been a paid subscriber of The Economist for about 8 years now, I have come to identify their bias as ‘Economically Conservative, Socially Liberal3 4.

Continue reading Bias is good

Tagging – Only Follow, No Filter

Hash Tag Conversations

For consumers of content tags, or #tags, serve two primary functions:

  1. The Follow function: To follow news of interest (e.g. a column with #lbl tweets showing latest updates from the race without me having to follow it live on TV), and
  2. The Filter function: To filter out specific content from the stream for various reasons, such as
    1. to either avoid listening news before we want to (match scores, movie spoilers), or
    2. to avoid getting drowned in updates during big events (SXSW, Google IO, WWDC, IPL, SuperBowl tweets taking over the timeline for brief periods), or
    3. to remove news from the timeline that we’re not at all uninterested in.

Content & platform companies all love the follow function, and have tried to make it as easy as possible for users to access it.

It’s understandable. Apart from allowing easy search, this also presents a straightforward way of targeting advertising to users based on interests. This ability to show relevant advertising – Specialized Shivs to users following IronMan world championships, and the latest BAAS to developers following Google IO – is extremely valuable to these companies.

The filter function, on the other hand, is almost universally neglected. None of the content consumption platforms that I use –  Twitter, Google+, WordPress and Pocket – offer any easy built-in way of filtering out content. All of them make it trivially easy to follow specifically-tagged content using tags or #tags.

A large number of popular 3rd party Twitter clients have the feature to filter out specific content, indicating the strength of user demand for the filter function. That 3rd party clients have this feature, also indicates that technical complexity isn’t the reason holding back content platforms themselves from providing this function.

Users want to cut out noise & irrelevant info from their content streams, yet none of the content serving companies make it easy for them.

Are there any technical, UX, business, or legal reasons for most content companies not providing filtering functions, or is it just a conscious, unfortunate, neglect of end user needs?

I’d really like to know.

TV News – unfiltered good, taxing bad.

bbc news logo

I like following the news on TV, along with the web, for primarily one reason – lack of a filter bubble.

When I watch channel 503 on Sky, I’m hearing the same news as anyone who’s watching the channel in, at least, London and the South East. Very unlike my Google News page, which I’ve set up so articles on my interest areas to turn up on top, and to show as few articles as possible from Telegraph (too right-leaning), Guardian (too left-leaning) and FT (requires sign-in to read), and more articles from The Independent, The Economist (I have a subscription), and The NYT. My only other source of un-filtered news, other than TV News, is the daily Quartz news email.

While I love the unfiltered nature of TV news for keeping me updated with happenings in areas, it can also be very wasteful. It’s here that I’d like some tweaking. For instance, my ideal BBC News (or CNN, which I also watch occasionally) bulletin should still cover all the news stories that they are covering for everyone else in the London & SE. However, just marginally personalise it for me by, say, covering Ukraine more, MH 370 less, Labour’s Europe plans more, politicians’ comments about Bob Crow less, Cycling news more, Cricket news less, and, most importantly, Silicon Roundabout more, Oscar Pistorius trial much less.

The formula for my perfect new channel/app is simple: Keep giving me *all* the news. Keep giving me the same tone & message of news that everyone else is getting. However, modify the volume of coverage devoted to each news item based on my interests.

Anyone out there working on this? Let’s talk!

Continue reading TV News – unfiltered good, taxing bad.

Company-watch on reddit

reddit-logoPro-tip for marketers:

Learn what people are saying about your company on @reddit with a simple URL:

http://www.reddit.com/domain/[your-domain]

Example: Track any posts on reddit mentioning/linking Europe’s leading seed investment program here:  http://www.reddit.com/domain/seedcamp.com

P.S.: If the scarcity of posts on that Seedcamp mentions page bothers you, get in touch with the program here: Wanted: Head of Content.

TV news filter bubble

US News Channels Prime Time Viewership

Quick thought from this chart in The Economist:

CNN was the flag bearer for middle-ground, relatively neutral news gathering in the US, and increased polarisation in US politics after Obama’s election seems to have hurt it the most.

The chart seems to indicate that CNN lost viewers to Fox News in 2008-09. This seems to line up well with the post-Obama election backlash from conservative right, and rise of the Palinistas-Ryanistas / Tea Party movement championed by Limbaugh and supported+tapped by Ailes.

2009-10 saw CNN losing market-share again, while MSNBC held steady – must have been about the time when MSNBC started tilting its programming towards the Democratic left1.

Net result: In 2 years, CNN lost half its prime-time news audience, as viewers on the right and left moved to news networks that echoed their own political views.

While researchers worried about effects of filter bubbles on the internet, the offline news consumption in living rooms was being filtered just as fast.

Read the full article on CNN’s recent transformation here.
Continue reading TV news filter bubble