How is Agile like a frog in a pot of water?

April 29, 2011
Frog example

Image via Wikipedia

Remember the old tale about the frog and a pot of water?   In essence, if you put a frog into a pot of very hot water, it will immediately jump out.   However, if you put a frog into a pot of cold water and gradually heat it, the frog will allow itself to be cooked to death.

Not to worry — I have never done either of these and I don’t suggest that you do so either.

However, there are lessons from this tale that I believe apply to the Agile development methodology…

I believe that a “successful product” must provide a valuable resolution to one or more needs from your target market.  Otherwise, nobody will buy it.  So when I talk to other Agile practitioners, I am amazed at how little thought they give to getting accurate and reliable feedback from their market.  Many of them make one (or both) of the following mistakes:

Product Owners Focused Solely on Development

The product owners of some companies spend so much time with Development that they quickly lose touch with their market.  Or even worse, allow the product direction to be set by the 1 or 2 customers that they have time to talk to.   Sure, they could talk to a number of customers via email, Skype, Twitter, etc — but you will only get answers to questions that you ask or reactions to significant positive (or negative) aspects of your products.  You may never know that some function takes twice as many clicks to complete as it should or that users are constantly using other tools to compensate for functional gaps in your product.

To know what your customers really “need”, you have to talk to enough of them so that you understand the difference between what a few customers say they want and what your product needs to provide to satisfy your target market.

Note:  If you’re interested in how to do this, consider this article written by Steve Johnson, Luke Hohmann and Rich Mironov  published by Pragmatic Marketing,

Developers Gone Native

Some companies encourage Developers to engage with customers so that they can get first-hand knowledge of what the customers need.    They carefully select the Developers for this task, because some are simply better suited for this type of interaction than others.

However, there are at least 2 problems with this approach…

  1. The Developers can’t spend a lot of time talking to customers — because they are typically senior people who need to be contributing to the product, and
  2. The Developers usually do not have the skills / experience to identify what each customer really “needs” (as opposed to what they say they want) and to synthesize the requirements for the product’s target market from feedback from only a few customers.

As a result, the Developers often focus on providing the capabilities that the customers they have met say they need — almost as if they were working directly (or “natively”) for those customers — instead of providing capabilities that benefit the majority of the product’s target market.

The Frog and the Water

If done right, Agile allows you to get rapid feedback from your market to keep you on course toward creating a successful product.  But your product can “go off the rails” just as easily with Agile if you don’t understand how to obtain and utilize customer feedback.

Waterfall is like the pot of hot water — it provides dramatic evidence of how well (or poorly) your product matches the needs of the market.

Agile is like the pot of cold water — and each iteration increases the temperature of the water in the pot a bit.   So if you do not have good people who are experienced in obtaining and understanding market feedback — you could, like the frog, sit in your pot until it is too late.

What Should You Do?

I believe that each product development team should consist of people who are highly qualified to do their specific jobs.  I look for teams where well-trained engineers are developing a product that is being tested by experienced QA personnel.  I’m not fond of teams where “everyone does everything” — because not everyone can be an expert at everything.

Likewise, your development team should include someone who understands the “art” of obtaining feedback from current (and potential) customers and synthesizing that into prioritized product requirements.  Whether you call this person a “product owner”, “product manager” or something else is up to you.   But they need to spend time at least 25% of their time with your current (and potential) customers.  Otherwise, they will quickly lose touch with what your target market needs and will have to rely on their own opinions, guesses, and/or hopes.

Creating a software product is very expensive and you can’t afford to make very many mistakes, because in many cases your competition is only “one click away” for a current (or potential) customer.  So avoid the temptation to ask developers to do this job “in their spare time” and get yourself a good product manager who can do this job “full time”.

Advertisements

Which mobile platforms will businesses embrace?

July 31, 2010

Each day it seems like there are 10 new applications for your iPhone, iPad, Droid-phone, etc.  And as you might expect, Windows-based tablets, according to Steve Balmer of Microsoft in a recent CNET article, are due out later this year (2010) with a bigger push early next year.

Then there is the continued growth of e-readers from a variety of sources such as Amazon and Barnes and Noble, with some people projecting that they will replace paperback books in the near future.

And while each new platform may be attractive to one or more market segments of consumers — the real opportunity may be the level of adoption of a platform by businesses.

If you are a company like Airbus, Boeing, John Deere, Whirlpool or any other large manufacturing company, you have many thousands of employees and suppliers.  You depend on digital devices to communicate product design information and to facilitate the collaboration of your personnel — especially across geographies.   Today, the primary device is a personal computer, usually a Windows-based laptop.

But as we all know, laptops have their limitations and may be overkill in a number of situations.  For example, if you want to know whether a part is positioned correctly inside an airplane fuselage, you may want to refer to a digital image of the drawing (or 3-D model).  Trying to hold you laptop in one hand while you position the part with the other hand is awkward at best, especially if you have a large laptop.

What type of device is best (and most cost-effective) for this type of situation?

A laptop?  An iPad?  An e-reader?  Or perhaps just a Smartphone?   Many people would argue that “any of these might work, depending on the situation”.

However, since sharing digital data within a corporate environment requires secure high-speed communications, I believe the bigger question is “How many of these digital platforms will businesses decide to support?”

And given the cost of deploying internal networks with ever-increasing bandwidth and the costs of distributing and maintaining each platform — I believe that devices that can easily connect to existing internal networks and can quickly download content from existing servers through web-based clients are most likely to be adopted by companies.

In other words, I think that the corporate “hand-held device” market could easily be dominated by Microsoft — IF they are able to produce high-quality, high-performing, and easy-to-use products and get them to market in sufficient quantities in a timely manner.

What do you think?


New markets for the Apple iPad?

April 8, 2010

The Apple iPad is finally here.  And the first batch of die-hard Apple fans have waited in line for hours, purchased them, and hurried home.  The blogs are simply overflowing with comments.

But I think the iPad’s impact in business applications will be much more far-reaching.

Today, there are a wide variety of hand-held applications in use in a wide variety of industries.

For instance, some doctors use laptops connected to wireless networks in their office to reference and update their patient records.  Other doctors use Palm, Windows-mobile, Blackberry, or even iPhone devices to review information, record observations, and order treatments, tests, and/or medications.  Patients are also issued similar devices to collect, record, and report treatment results.

In a similar manner, discrete manufacturers often use personal computers or hand-held devices to provide instructions (including images), to manufacturing personnel, technicians and the like.

This means that every software provider that wants to offer solutions to these industries has to develop, test, and deploy their products for each one of these devices.  They also have to develop, test, and deploy server-side components that are usually unique to each hand-held device.   It is hard to even list all of the testing permutations that each of these software providers must currently endure, especially if you take into account the various operating systems versions that must be supported for each hand-held device.

Into this environment comes the iPad.  If it is just another device, then each of these software providers must adopt yet another device and devote even more development, testing, and deployment resources.

But what if the iPad can achieve enough popularity with businesses so that it replaces one or more of the current hand-held devices?   What if users become so thrilled with the iPad’s large touch screen user experience so that they don’t want to use any of the other devices?

Imagine how much more productive these software companies could be if they only had to develop, test, and deploy a single version of their application — for the iPad.

I’m not an Apple fan in general, in fact I’m writing this post on a Microsoft laptop and use a Windows mobile phone.  However, I’m a big fan of software products and recognize that if you have fewer platforms to support, you should be able to get better products to market even faster.

But on the other hand, I expect a lot of iPad-like devices in the near future just as there are now a number of iPod-like devices, which will make the lives of these software companies even more complicated.   I just hope that the end result is a much better user experience for the users of these devices.


12 Social Media Secrets – Part 2

April 8, 2010

The good thing about going to the dentist is that you have time to think.   You can’t talk much — not with someone’s hand in your mouth.  And they usually don’t know you all that well — so they don’t have much to say to you either.  So you are left to your own thoughts.

In my case, that meant a return to my earlier post about the 12 Social Media Secrets, because I realized that I hadn’t talked enough about the types of tools and information that I would like to see.

12 Social Media Secrets – Part 1

I believe that the usage of social media in the next 5-10 years will explode to a level that few of us anticipate.  Thus,  it will be common for a company to get thousands upon thousands of inputs (tweets, emails, blog comments, etc.) about its products on a daily basis in the not too distant future.  If you doubt this, consider how many text messages people around you are sending and receiving each day.  Five years ago, text messaging within the US was not all that common.  Now, we have legislation pending or in place that forbids texting while driving because too many people simply “must” text at every minute of their day and unlimited text messaging plans are considered commonplace.   The data flood is coming, and sooner than we think.  (Some would argue that it is here already…)

Therefore, we are going to need tools to process all of the inputs we receive from customers (and prospects) about our product and I, for one, want to be sure that I’m getting good information from all of the market segments that I care about.  I can’t afford to be misled by a few, vocal participants or to ignore a particular market segment that might be critical to the success of my product.

First, I’d like to know as much as I can about the people behind the inputs.  This includes:

  • Gender (male or female)
  • Age Bracket (e.g., less that 21, 21-30, 31-40, etc.)
  • Marital status
  • Employment status (Full-time, Part-Time, Retired, Unemployed)
  • Country where  they live
  • Closest city
  • Number of people within 10 miles of their house

Other information like race, primary language, household income, number of children, etc. would also be useful — but is more controversial and therefore much less likely to be available.    The goal is to ask for information that is anonymous enough that people will feel comfortable providing it while also gathering enough information so that you can make intelligent decisions.

And besides, if I know that you live outside of Boston (like I do) and that there are only about 35,000 people within 10 miles of your house — I can guess that you live in a medium-sized town in one of the Boston suburbs.   And from that I can approximate the medium income for people like you.  It won’t be exact, because some Boston suburbs have much higher medium incomes that others, but it might be close enough.

However, for this information to be useful, it needs to be consistent across all of the social media tools that my customers are likely to use and kept in a single location so that users don’t have to run around updating multiple sites.  A Gravatar is a good start, but it needs more information.

Second, I’d like to be able to analyze the quantity of the inputs I receive.  For example:

  • Number of inputs ( Tweets, emails, blog comments, etc.) received from the same user in a given period.
  • Number of inputs from the same market segment (gender, age bracket, marital status, employment status, geographical area, etc) in a given period.
  • Average number of inputs for each market segment during a given period.
  • Market segments that didn’t provide an “average” number of inputs during a given period.

Third, I’d like to have an application that can automatically organize the inputs I receive.  For example, I’d like to be able to distinguish between an input from a dissatisfied customer and one from a potential new customer.  Of course, I can establish numerous different input channels — but there will always be an input that is in the wrong channel.

I’d also like the ability to aggregate all of the inputs that I receive — so that I don’t have to manually organize my Tweets and then cross-reference them with emails received from similar people in similar geographies.

Lastly, I need tools to allow me to quickly organize and manage all of the inputs that I receive.  For example, that dissatisfied customer who complains each day that their cell phone is difficult to use because the screen is impossible to read in bright sunlight has a very valid point.  However, I don’t want to read the same complaint over and over again for weeks on end — I won’t have the time.  I need to be able to manage all of my inputs effectively and efficiently — so that I can gather the key messages very quickly and then spend the rest of my day improving existing products or creating new ones.

I see lots of opportunities for this approach within social media — but not a lot of tools that provide anything close to what I’m looking for — at least not yet.


12 Social Media Secrets – Part 1

April 8, 2010

I read with interest the blog entry “12 Social Media Secrets From Worlds’ Top Superstars” from the Social Media Examiner (link below).

http://www.socialmediaexaminer.com/12-social-media-secrets-from-worlds-top-superstars/

Some of their suggestions make really good sense.  For example, it’s always a good idea to engage your audience and then listen to what they have to say.  And given the diverse nature of our global economies, social media may be the only method by which you have any hope of getting reasonable feedback from your target market in any reasonable time frame.

However, I’m still bothered by some aspects of Social Media.

For example, does it make sense to repeat something over and over again?  (See #11 “Repeat Your Tweets”)   The idea is to repeat your Tweets so that people who didn’t see them the first time will see them the second or third time.    However, what about those people who saw them the first time?   Are you going to turn them off because they don’t see anything fresh and new from you? Or are the people who saw your original Tweet “better” customers (or potential customers) because they were following you close enough to see your post the first time?

Looks like we also need better filtering and prioritization capabilities in Twitter so that we can more easily process the hundreds (or more) Tweets that we get each day.  I don’t know about you, but I would really like the ability to organize my incoming Tweets into subject “folders” and then quickly see a summary of how many Tweets each folder contained so I could prioritize my reading.  Imagine being able to distinguish media reports on your company (or products) from those provided by current customers and from inquiries from prospective customers.   That would be powerful.  I hope someone at Twitter is listening…

I also wonder about the quality of the feedback that you get from Social Media because it is almost too anonymous.  If your product is targeted at “married, middle-age men in the United States”, how can you determine whether the feedback you are getting is from your target market or not? You could analyze the Facebook, Twitter, and LinkedIn profiles of everyone who responds, but no company has sufficient staff to do this unless you only received a few responses.   And if you only got a few responses, you probably want to re-think what you are saying and how you are sharing it — because you’re not getting enough mind share.

Another concerning aspect of Social Media is that the feedback you receive can be dominated by a small, highly vocal segment of your target market.  I’ve written about this before (link below).

How can you use social media to interact with ALL of your target market members?

However, with all of the buzz about social media it was easy to think that I was worrying too much or jousting at shadows.  Thus, I found the last “social media secret” to be perhaps the most important, where Jason Falls says:

“Social media success depends upon your type of audience, product, company, network and environment.  You need to know your brand, your audience, how to communicate within each specific social network or online community and the right tools to use.  Only the last of those is fairly predictable knowledge.”

Seems to me that social media is the best tool set available for getting feedback from your target market, but like any tool it has its strengths and weaknesses.   Thus, in typical product management fashion, I keep asking:  What could (or should) we do to improve these tools?


Is it possible to detect trade secrets disemminated through social media?

April 1, 2010

I read with interest the announcement of the coming Boston Events on social media and how to protect your products, brand and reputation.  (It is scheduled for April 7 at 8am.)

Protecting Your Brand, Reputation, and Secrets in the Age of Social Media

One of the key topics in this session involves preventing your trade secrets from being disclosed through social media.

This got me thinking — How could you identify one of your trade secrets that was distributed through social media?

In today’s global economy where products are often manufactured by a company’s supply chain partners, your ability to protect your trade secrets is absolutely critical to ensuring the continued success, and even existence, of your company.  Otherwise, the highly innovative product that you finally got to market after years of research and development could be cloned and mass-produced by a foreign competitor without the same R&D costs in 90 days (or less), effectively destroying your ability to generate the revenue you were expecting.

Obviously, you could manufacture the most proprietary elements yourself and provide all of your employees with phones, laptop computers, etc. so that you can legally search all of their outgoing emails, chats, text messages, etc. for relevant trade secret content.  But no company can afford this level of investment or can dedicate the staff necessary to perform all of this “policing”.  And few employees would tolerate the continual monitoring of their professional and personal communications that this approach implies.  After all, they are not working for the CIA or someplace similar where they expect this sort of monitoring as a condition of employment.

Or you could carefully restrict access to any trade secrets to only a select few employees.  However, this approach limits the ability of “average employees” to identify product or process improvements that have saved other companies millions of dollars in product development and manufacturing costs during recent years.

Thus, if you are going to actively detect a trade secret violation, you need to be able to deploy “detection” technology that can identify trade secrets being compromised and alert you accordingly.

But is this really possible?

Granted, searching text-based content for trade secret information on Twitter, Facebook, Blogs, etc. is fairly easy using Google and similar tools.   However, none of these tools are good enough to capture violations that don’t include relevant key words.

For example, you decide that the phrase “glass half full” is a trade secret and set up automatic filtering on all out-bound text content to detect any violations.  However, a message that talks about a “drinking container which has an equal amount of liquid and air” is, by definition “half full” but is unlikely to be identified by your automatic filtering tools.

The problem is even worse for videos.   Google, Brightcove, Wistia, and others have tools for identifying and tracking the usage of videos.  But all of the “searching” algorithms I have seen depend on the title or the meta tags applied to the video. I haven’t seen anything yet that could parse a video for a given text string or word — even though a number of companies have done research in this area.  Presumably the NSA has huge banks of super computers somewhere that can do some of this — but this is only a guess — and even if they have this capability, it doesn’t help the average company.

And since a potential violator is unlikely to post a new YouTube video with a name or tag that says “trade secrets of company xyz”, this seems like an unreliable approach to me also.

So the question is — Is it truly possible to detect trade secrets being disseminated through social media?   Or are we forced to rely solely on the trust worthiness of our employees,  colleagues, and business partners?

And how can we trust our supply chain members, when we often do not even know which companies are providing key elements to our key supply chain members?


Do you believe in Gemvara’s (Paragon Lake) strategy of customized products for consumers?

March 26, 2010

Gemvara (previously known as Paragon Lake) is focused on providing its customers with customized jewelry.  In other words if you wanted a new pair of earrings for your wife or girlfriend, you would select the setting, the material (gold, silver, etc.), and the stone(s) through their website.  Your jewelry would be custom produced for you and arrive in the mail.

Originally, the idea was to have their service available through jewelry stores — such as those that you see in every mall.  Now, they are focused on selling directly to consumers — a move which has resulted in some changes to their organizational structure.  (Story details below)

Gemvara CEO Deb Besemer steps aside, due to new e-commerce strategy – http://bit.ly/d67Xy6

I wonder about this strategy because jewelry purchases are very personal to me and I could not imagine buying something like this over the web.  I just wouldn’t be able to see the finished piece nor let my wife try it on to make sure it was what I (I mean, she) wanted.

But the bigger question is — Are US consumers now ready to purchase customized products over the web and wait for them to arrive?

Years ago, Rover Cars of the UK, wanted to offer its customers the ability to walk into a show room and order their customized automobile.  You would be able to pick the model, exterior color, interior color, numerous features, etc.  The car would be produced to your specifications and arrive at your local dealer in less than a week.  Unfortunately, Rover wasn’t able to sort out all of the manufacturing logistics at that time, so the effort was shelved.    Subsequently, other European car manufacturers have implemented similar capabilities so that most Europeans order their custom cars and wait for them to arrive — a big difference from the US approach where you select the model which is closest to what you really want from a large inventory on the lot and drive the car home almost immediately.

If Gemvara is correct and US consumers are ready for this type of buying experience, the potential impact to product design systems (so that you can account for all of the many product option combinations), manufacturing processes (so that you can produce customized products on-demand), shipping, customer call centers (to resolve complaints), and return policies (so that you can handle customers who didn’t like what they received) are immense.

And let’s not forget all of the systems that will need to help each customer configure, visualize, order, and pay for their products or the computing infrastructure that will be required.

Who knows, you may finally be able to buy a car in the US without having to endure all those hours spent negotiating with your sales rep, his manager, and whomever else the dealership injects into the process.


%d bloggers like this: