Archives For Apps



Whatsapp has successfully capitalized on the messaging market by changing some old habits around SMS and chat applications.  Slack now is attempting to do the same with email.  In a way, we could look at Facebook and say it is just one giant mailing list after all!  And yet, when we talk to the users, they find their experiences with Facebook or Whatsapp to be “different”.  One of my friends the other day argued that sending something on Whatsapp is a lot easier than sending something over SMS.  When we parse this step for step, it only saves steps when the messages are to a group, since Whatsapp has a persistent notion of groups. When pushed to explain, my friend was unable to really say why.  I’m sure all of us have found that sharing a photo to Facebook is a lot easier than sharing it over email.  Once again, quantifying this is hard.

I have recently become a Slack user and I find it significantly easier than email threads.  It’s still early days to draw conclusions on this, but so far, it seems to be simplifying my collaborations.  One could argue that conceptually, it is not really all that different from discussion boards or forums or even email threads at the core of it.

So, why are these new world apps doing so fantastically well on age old problems?  It is really important to understand the difference between creating products that have “differences” from existing products and creating those that are just…, well, “different“! In fact, you need the perception of being different!

In his book, “Hooked”, Nir Eyal states that one way of creating successful products and changing people’s old habits is by taking a problem, creating a solution and then taking away steps from it until you are left with just the bare minimum steps to solve the problem.  Let’s take the case of sharing a status over Facebook, for example.  You are able to write something and post it – there is no need to select a mailing list or a group to share stuff with, let alone handle the creation and management of such aspects.  A select few will whine about how they don’t want the entire world (which seems to be roughly equivalent to the set of friends we have on Facebook these days) to know about their status.  In reality though, algorithms and the users have been getting smarter at this – algorithms figure out how to bring you the key updates of relevance to you and the users are getting better at tuning out what they don’t really want to see.  It is really the same thing that Whatsapp has done with messaging.

The lesson here is very simple and yet hard to do – revolutionary products do not always need green field problems.  Revolution in products can simply be about taking an existing problem and approaching it from a user centric perspective.  Building products with incremental differences, especially in a field where some habits have been established, is not going to be useful.  However, showing that your product is different (in a simpler way!), even if it were only in a single (but important) dimension, could make all the difference!

Disclaimer: All thoughts are my own and does not reflect the views of my employer in any way. 


Telegram saw 8M downloads the day Whatsapp was acquired by Facebook.  This is not new.  When Parse was acquired by Facebook, the blogs rushed to write about why this is great for their rivals. Stackmob accelerated its Parse migration pipeline and came out with it in just a weekend! There was outrage when Tumblr was acquired by Yahoo!.


There are, of course, several reasons for such reactions and in each case, it is slightly different. Early adopters get so entrenched in their favorite platforms that there is a sense of ownership – when a drastic change occurs, it feels like their trust has been misplaced or that they have been betrayed.  But, beyond all this, there is another challenge here that we are seeing, such as in the case of the Whatsapp-Facebook situation – why should Facebook have all my data?  This alone causes a split market in terms of data ownership.

When Facebook published its recent upgrade to the Android app, it asked for permission to read SMS.  As much as I like to be on top of the world of mobile apps, I said no to the upgrade on my primary phone.  I had enough secondary devices on which I don’t use SMS to try out the new app!  To this day, my Facebook remains at v3.9 on my primary phone!  The thought of Facebook reading my text messages just did not sit well with me.  Of course, now I’m faced with the challenge of using or not using Whatsapp!  (Just to remove any ambiguity, I fully plan on continuing to use Whatsapp, unless Facebook decides to mess with it like LinkedIn did with Pulse!).


Recently, a friend that I recommended SwiftKey to said that he did not agree to SwiftKey learning from his GMail – they had no business knowing the content of his emails!  My reasoning around the benefits of personalization that can shave off minutes in typing a single email did not manage to convince him.

So, what exactly is behind these strong feelings about who can or cannot read the various parts of our data?  Mostly, just personal principles.  For most people, when it comes down to it, as long as the data is “secure” and “private”, this means nothing and they only stand to benefit from all the personalization it can enable.  However, there are two problems – we don’t always believe it is in fact, “secure” and “private” and we have our biases in which companies we love and trust.

But the knee jerk reaction to these acquisitions tells a very interesting story.  In reality, we are faced with this particular challenge:

Do I want to give more of my data to the bigger companies that can aggregate various types of data to learn all kinds of crazy things about me? Or, do I want to give my data to a small startup that has no resources to even consider implementing security correctly? 

This is a very difficult conundrum, particularly because, “implementing security correctly” is a non-trivial task, that most developers are quite bad at by default.  When you are big, you have a responsibility to keep the data secure – way more so than we can imagine.  When you are small, there is no real upside to spending the time on security.  It slows down the development to think about it from an architectural perspective and get the pieces right. All the security holes in Snapchat and other small apps are testimony to this. Security gaps happen even in big companies, where this is taken seriously and experts are hired to ensure correctness. One can imagine why it is more or less just “winged” in the smaller ones. This is not a reflection of anything in particular – it is often just a lack of resources to focus on everything, when you are a startup.

I don’t particularly have an answer to this conundrum.  But, as a user, convenience trumps everything – which means that I will use an app from a small startup if it does the right things to make my life simpler.  That said, I generally have less issues with giving up my privacy to the bigger companies – the value that personalization can bring is huge and I’m looking forward to it!

Twitter brings curation and quick summaries together.  In an information overloaded world, that is powerful.  But until they can prove they understand user experience, it is hard for me to take them seriously.  


Twitter’s IPO has been big news.  After all the speculation on their valuation and the criticism of not having that critical woman board member, a 63% spike on their opening price was not a bad show!  26% of teenagers think Twitter is an important social network.  Evidence suggests that people are more likely to follow influencers on Twitter than on blogs or any other places.  Snapchat, the ephemeral photo exchange app, is valued at least at $3B.  The average age of Facebook users is going up and the younger generation is migrating to the cooler places – Twitter, Vine, Snapchat, etc.

Twitter has taken short communications to the mainstream in a massive way.  SMS was always fairly popular – but, Twitter took it to new levels of popularity by providing equivalent functionality, only richer in content!  Other apps such as Whatsapp and Snapchat have followed suit in a similar vision of short messages, but branching in the type and mode of content exchanged – and in Snapchat’s case – limiting the time to live for a piece of content.

While there are a number of possible explanations for the wildly growing popularity of this style of messaging, one that I think is a major contributor is human attention span.  Variety is interesting.  Holding our attention span for long on one topic is hard.  Topics get boring.  Just as the tide was turning from theres-a-lot-of-information-to-catch-up-on to struggling-to-keep-up-with-the-information-pace-and-volume, these short messaging innovations caught up with us.  The illusion of being able to catch up with information quickly is attractive.  Being able to quickly produce content also helps – a single picture or a few words can get it out there.

Curation combined with short messages surely allow us to see a preview of information, leaving it for us to decide whether we want to consume more.  Of course, this is simply hiding information behind yet another level of indirection – a typical computer science solution to problems.  The real content is buried in links that are increasingly shared as these short messages.  These links often lead us to more old style “blogs” (I’m guilty as charged!).

Bringing curation and quick summaries together is clearly the strength of the Twitter class of platforms.

Yet, taking Twitter seriously is tough for me.  Why so?  Fundamentally because they are yet to prove they understand user experience. Reading the Twitter stream on the Twitter app on the phone is painful.  There is nothing that screams “come spend time on me” on this interface!  If you want to see a worse design of a new generation app, you can take a look at Quora, but, we’ll stay on Twitter for now.


The highlighted stuff provides zero semantically useful information. The user name gets a bold typeface, but then, the picture already tells me who the user is!  The rest of the text is all uniform, resulting in a massive stream of text on the screen!

This is why it is hard for me to take Twitter seriously.  The real brilliance in next generation content sharing is going to be two-fold – semantic information extraction and presentation.  At one glance, I should be able to extract the most meaningful summary of the content I’m trying to consume.  Once this happens, the need for platform level indirection (i.e., Twitter leading to TechCrunch) decreases – rather, the summary can come directly from the content provider.  Although, as innovation goes, it is unlikely that it will come from the content provider and hence, some platform that summarizes and presents (note that it doesn’t have to be the same one doing both) will likely evolve.

Could that be Twitter in the future? It will certainly be great shareholder value if Twitter can figure that out!  But until then, I will continue reading my tweets on Flipboard, ignoring the full page Twitter app ad that now regularly appears in my Flipboard stream.  After all, flipping over it only takes a second!


The tech community has demonstrated that it is at best confused about what type of diversity businesses should aim for, in order to advance well.  It is one thing to curb discrimination, but on the topic of diversity, I think we should be looking for “diversity in perspectives”.  

Diversity has hardly been an easy topic to understand.  For centuries, the human civilization has been trying to deal with just what diversity is and how best to handle it.  So, it is no surprise that it leads to exhaustion as Dick Costolo gets hammered on the lack of women on the Twitter board.  Those who followed the NYTimes article and the less than ideal exchange that happened on Twitter following a harsh comment by @dickc know that the situation is getting so much attention.

Screen Shot 2013-10-08 at 11.58.05 AM

On the topic of “women as minority”, I have very conflicted views.  I’ve written earlier on this topic, advocating that women need to be confident and trust in their abilities to be as good or better as their male counterparts.  So, this question of “should an organization be compelled to have a woman on their board” makes me very uncomfortable.  If you are a woman, would you like to be hired because you are a woman (of course you’d be subject to a minimum bar) or would you like to be hired because there isn’t a better candidate than you for the job?  I’d prefer the latter myself.

For what it’s worth, I think we are hung up on the wrong question!  We are implicitly making this be about discrimination rather than diversity in a meaningful sense.  Have you hired the best candidate you could find for the job?  Of course, there are all kinds of other factors – a reasonable timeline for one – but, in general, if you answered yes to that question, you are done!

Diversity In Perspectives

Now, diversity in perspectives, however, is a completely different thing.  A company like Twitter needs creativity in multiple dimensions.  Arguably, understanding female users is one area.  But, I have a hard time professing that you need a woman to understand female users.  Would you hire a teenager to the board so you can understand teenage users?  Not necessarily.  In a similar manner, the key is that you have a board that can bring in diverse perspectives that are important to your business.  The real problem lies in the fact that most leaders are not necessarily excellent judges of other people’s strengths.  So, it is generally hard for someone to understand who is bringing the right set of perspectives in which area.  How do you know that a male director you just hired understands female users?  It is much easier to believe that hiring a woman will bring that perspective.  All the noise about how Twitter has so many female users and hence desperately needs a woman on the board relate to exactly that!

It is true that it is often difficult for people to understand how their real users behave and what they need.  The solution to this is not that they go find themselves a representative from each segment of their user population.  First of all, there is no guarantee that a 40-year old woman can bring the perspective of a teenage girl.  So, unless you got your segments exactly right, that would not be an ideal fit.  Second, not only do you need the right perspectives from different user segments, you also need these people to be able to connect the dots together and create a cohesive strategy.  Not to mention be able to identify user segments of future relevance.

The bottomline is that diversity in perspectives is what is most important and you need creative people who can understand that they are not necessarily the representative user, can understand the actual users and project their future wants and needs and equally critically, can work together!


I’m a fan of LinkedIn.  I have watched that business grow and what they’ve accomplished is nothing short of amazing.  When they first came out, you thought “hmmm.. online resumes, cool!”.  Now it is the place for anything remotely related to professional activities.  It is where employers go to find potential candidates, entrepreneurs connect with each other, people increasingly go to get anything about their professional lives – from news about other professionals to general water-cooler-discussion-worthy news.

I often use LinkedIn as an example for how to successfully introduce the user to simple functionality and add functionality in bite sizes.  Once users get used to what they have, they’ve slowly and steadily added other things and educated the user on how to use new functionality.  It has worked rather well – okay, they’ve had to roll back on some stuff and double down on some other stuff.  They’ve had their own series of features they’ve phased out (the answer forum or publisher pages come to mind from recent days).  But, let’s talk about what appears to have worked out.

First you put your resume on LinkedIn.  Next, you connected with people you know.  Then you got introduced to people you want to know.  LinkedIn introduced jobs and subscriptions and got stronger ties between recruiters and possible candidates.  You got your next job because someone on LinkedIn saw your profile and contacted you.  LinkedIn pushed the privacy limits with public profiles and got you more visibility.  You got to see who’s interested in your profiles and it got addictive.  Recommendations were introduced to fuel professional introductions.  Fast forward to the current state and we now have endorsements, where we can proactively certify someone to be knowledgeable in something.  Here is an interesting visual history of LinkedIn.  This is all great – 225 million users in 200 countries is worth talking about!

LinkedIn has made it into my frequently visited sites and it is almost the only network, where I choose to receive updates on my primary email address.  By comparison, my Facebook, Twitter and other network emails go to an address I almost never check – I don’t get those emails on my phone and I rarely ever login to those accounts even on a laptop.  I go to them when I feel like it – at a pace that I feel is sufficient to keep up.

Lately, LinkedIn has been pushing its luck a bit too far.  First, the number of emails I get from LinkedIn has exploded.  I started having filters, but thanks to GMail’s new “social” classification, my Inbox is back to being sane!


But, more importantly, I’m bothered by the amount of real estate that LinkedIn thinks I should get on my screen for something I’m explicitly looking for.  The screenshot above probably explains itself.  I think LinkedIn endorsements is a great idea.  It is not mature right now and in order for this feature to bring value to employers, it has ways to go.  But, it is definitely in the right direction.  That said though, does it have to be in my face every time I try to look up someone’s profile?

That screenshot is the profile page of someone I pulled up – when it comes up, the actual information I sought has the least real estate on that screen!  Seriously?  Between endorsements, ads and other recommendations that LinkedIn wants to throw at me, the profile I actively sought out gets a fourth or less of the available real estate! Thankfully, it doesn’t do that on the phone yet – but, where is this going?

I find this extremely annoying and if this continues, pretty soon, I’ll be looking up less people on their site.  I’m very curious to know if data points to this strategy being useful in terms of user experience as well as user actions (do more users provide endorsements because it’s in the face like this?).

This is the age of continuous experimentation.  But, I wonder, just how much it is okay to push the users?  I think there is a point in user acquisition where the barrier to entry for a new comer as well as the cost of quitting for the user are both rather high.  And for the most part, this is what brings complacence to LinkedIn or Facebook or anyone else in that state.

But, I’d have to imagine that user experience is still the top priority for these companies.  So, I have to believe that either data points to these abominations still producing good user experiences or that the metrics by which user experience is measured are all messed up.  But, more on that later.  For now, I’m not really sure how long I will still visit LinkedIn “frequently”.  My use of Pulse (now part of LinkedIn) rapidly declined after they shoved the Highlight feature on me, forcing me to highlight every time I wanted to share something.  My tolerance for such unwanted stuff is low – especially when the service isn’t indispensable to me… So, I guess we’ll see if I do really feel LinkedIn is indispensable!

Great product management is about what you manage to keep out of your product. 


I’ve certainly been there and I’ve watched most PMs do this – pack as many features as possible into a product or a release.  More features = better, right?  That provides the perception that we are capable of producing tons of stuff.  Of course, the better PMs will enforce quality bars on those features.  There will simply be a large number of reasonably good features that work reasonably well.

Except it’s all wrong.  The best PMs are constantly thinking about how to keep the product minimal.  It is what we keep out of the product that really defines good program management.  It is extremely hard to do and even harder to keep doing on an ongoing basis.  But, the best ones know that even at the cost of taking a gamble at going after the wrong feature set, that is the right thing to do.

I learnt this the hard way.  By trying to explain why there is nothing specific about Bluetooth or WiFi in application level peer-to-peer in a company that doesn’t quite understand applications.  And that sharing photos using peer-to-peer to one device in front of you and sharing it with 10 or 10,000 users across the globe requires the same fundamental underlying technology.  Sure, the techno geeks who lived in the same world as me got it.  But, I was unsuccessful in getting too many others around me to see that reality.  Instead, a minimalistic product needed to have been built, albeit with the grander vision influencing the architecture and design, to have created interest and momentum and avoided overwhelming the people who needed to fund the effort.  Luckily, this is an area where I’ve learnt from mistakes.

Recently, I’ve had a plethora of phone problems (minor digression and rant – my Galaxy Nexus has more or less died, I’m desperately waiting for Verizon to carry the HTC One and in the meantime, I have a completely unstable phone that drives me insane!).  In this state, the one application that has been a source of problem is Whatsapp.  Every time I move my SIM to another phone, it wants the phone number to be reverified and the setup to start all over again. While it migrates the groups I have created to the new phone, the messages do not migrate.  And, on the phone with no service, it constantly pops up the message that Whatsapp can only be installed on a phone with cellular service.  In fact, a few months ago, I realized that I couldn’t download Whatsapp on my iPod Touch.

Whatsapp is an application that works over data (ala IP).  There is nothing that technically prevents it from working just over WiFi (and hence, on any connected device).  However, it is not supported in Whatsapp.  In fact, there is a short and sweet message on their FAQs that states simply this – “We currently do not support tablets, computers, or Wi-Fi only devices, and do not plan to do so in the foreseeable future.

So, here is a case, where extra engineering effort was potentially spent to ensure that an app that otherwise would just seamlessly work on all devices only works on phones with service.  Of course, people have bypassed this limitation.  But, that is for the geeks and not for regular people.

This limitation in Whatsapp strikes me as an extremely thought through product management decision.  They were after the SMS market.  They modeled it as close to the SMS functionality as possible, only sending messages was free.  So, they tied it to the only id that SMS is tied to – the phone number.  If you changed devices, your SMS messages did not carry over – they modeled after this too, with an exception only for groups.  Groups, by itself is almost the only enhancement beyond SMS (and is solving an important pain point).

While this artificial limitation is annoying to people like me, it made the app ultra simple and allowed them to market this to vast parts of the population with no further explanation than

“Simple. Personal.
Real Time Messaging.”


“WhatsApp is working on building a better SMS alternative.” as they say on LinkedIn.

It is really important to connect with the users at a level they can understand and this is what they did.  Of course, I think it is terrible UX to keep throwing messages at you when you are not even trying to use the app (on a phone where you have previously installed Whatsapp and removed service).  But, that’s such a rare use case that is not worth their time solving at this stage.

Bottomline is that they went for the minimal feature set that aligned with their goals and got users on board.  That’s brilliant product management.  One that is extremely difficult to do – all of us want “backup” features, things that we can fall back to, if our primary plan does not go well.  Worse still, most of us have a hard time picking the single most important feature – there’s always a list of important features.  But, being able to figure out that minimal set is what makes a difference between brilliant products and good products.

We should always be thinking of Minimum Viable Product from the “Minimum” perspective. “Viable” is also important, but that generally doesn’t require special efforts to think about.  Hopefully, you have already understood how your product is viable or you would not be working on it.  Staying minimal requires special efforts.  And it makes all the difference in building a great product!

In his Google I/O keynote, Larry Page said “Law can’t be right if it’s 50 years old. Like, it’s before the Internet.”  Some people were riled up about that.  When I read about FAA re-considering it’s 50-year old ban on using gadgets on the flights, it reminded me of what Larry said.  While obviously some values don’t necessarily change with age, for anything that involves technology, I contend that 50 years is eternity.  Things just don’t apply as they did before.

We really are in an age where technology changes so rapidly and we are at the cusp of where technology really is taking over human labor in a big way.  How do lawmakers keep pace with this and how do we define laws in this society? I’m sure all of us have heard of ridiculous stories where someone has been pulled over for touching their phone in the car.  Yet, a study shows that the biggest diversion comes from paying attention to children in the car!  Do we turn around and create laws that prevent us from interacting with our kids in the car?  There would be an outrage if that happened!

I don’t know the answer to how we should be defining laws.  Part of the problem is that we cannot have one rule for amending laws across the board.  In the case of lawmakers, we are talking about folks who are at least one step removed from the pace of the technological changes most of the time.  But as it turns out, the technologists also fuel some of these stale rules.  Take the use case of automatically turning off your phone when detecting you are on a plane, for example – this one has been making the rounds in context aware research cycles for a while.  It works on the assumption that turning off devices on a plane is a given.  When technologists go after applications that support the outdated laws, it is a bad signal to me.

We need to be better at creating our own destiny – and by ‘we’, I mean the tech community.  We need to be able to distinguish when technology needs to bridge some gaps when the gaps make no sense and need to be addressed more fundamentally.  To really push the boundaries, that must happen.  And to really innovate, we must push the boundaries.

Is 50-years an eternity? In the world of self driving cars and networks on balloons, it most certainly is!  This doesn’t mean we change our fundamental human values every decade.  But, it does mean that the technologists help the government and lawmakers understand the leaps that have been made at some intervals such that our constitutional rights and laws can be kept relevant.


I wrote about the undesirable shifting UIs earlier. Here is an alternate perspective on such UIs. There are times when a shifting UI has a purpose.  When it does, it takes away the attention of the user to a new menu or a popup for a reason.  The user was meant to be looking at the changed UI and it may even be relevant to the user.

News sites often do this type of animation.  As the reader reaches the end of an article, they would show a peek into the next article.  That is a highly relevant distraction – one that is worthy of the user’s attention.

So, there is a place for these transient UI elements.  It is not all bad or all good – the key is, as long as the UI matches the intent of the user or an action that is a natural step for the user, even popups can be good.  Random shifts that distract the user from what they want to look at are not cool!

As I’ve admitted before, my design skills are self-taught and often acquired by just looking carefully at the design elements of some of the best apps that are out there.  One thing that has been bothering me lately is the intuition behind shifting UI elements – or the fact that I don’t get it.  Take the new G+ layout, for instance.  Here is a view of the G+ layout while scrolling down:


The first one is the default view upon visiting the G+ site and the second is the view upon scrolling down a bit.  I’ve been thinking about what bugs me there and realized it is the shift in UI elements that occur.  A couple of things happen there – the home icon gets replaced by the G+ symbol (but they both do exactly the same thing!); and the notifications indicator moves from the top area to the bar below.  Both these need not actually shift at all – for e.g., the default look of that bar can be that of the second image above.  When this kind of a shift happens while scrolling, it gives the illusion that something has changed and draws attention to the shifting element.  When the user realizes that nothing has changed, it is quite annoying.  It just shifted my attention to an element for no apparent reason, causing me to lose my focus on whatever I was looking at earlier!

I take the example of the web interface here, but the same is true on the mobile.  In fact, it is more pronounced there.  Some apps bring up certain UI elements only while scrolling up or down and it is quite distracting.  Scrolling up on the G+ Android app brings up a bottom bar:


What is it about scrolling up that suggests a user intent of wanting to do any of those four actions in the bottom bar?

I’m not trying to pick on G+ here – there are other apps that do this as well.  Pulse did this for a while where scrolling up would bring up options to share a post.

When such appearances of UI elements are tied to specific actions (such as scrolling in a particular direction) that are not necessarily indicative of associated user intents, it is confusing and distracting.

UI elements, down to every last detail, must be carefully thought out.  Especially on the mobile screen, where such shifts have a high impact on user attention, it is crucial that we debate the details of the design to every last bit.

It is possible I don’t get the idea behind these shifting UI elements and I’m missing the point.  As a user though, I have found them pretty distracting and annoying.  And putting on my self-taught designer hat, I cannot seem to be able to justify the shifts.  I’d be much more forgiving of a shift that is intuitive – where it ties to something that in fact, is channeling a user intent.  However, in the cases I’ve seen so far, it appears to be promoting the important elements of the app, using scroll direction as an excuse to do so!

I like my Android device the way it is.  I did try out Facebook Home to get a feel for it, but I won’t be going back to it anytime soon!  You can read my full review on Quora here.  In brief, it introduces a delay in all the important things I’d use my phone for.  If you need to make an emergency call, forget it – you can like and comment on Facebook news feeds while your phone is locked, but getting to the dialer for an emergency call will take a few steps! A number of the normal Facebook frustrations on the mobile are now available on all our apps!


All in all, I uninstalled after giving it a good shot and trying out various operations.  I’m all for a contextual future – but, my contextual envelope happens to be beyond Facebook!