November 2020 – Looking Forward

November 19, 2020

Straight Lines And Hockey Sticks On The Road To A Cash Crunch

As a member of a couple of angel investor networks, a former software executive, and a teacher of a graduate course on new product/service creation, I have seen many financial projections from startup founders or even new product managers in large companies.

One very common pattern for sales projections is a straight rising line (simple linear trend). Here’s one that shows consistent growth in sales, with expenses following along in a similar path. Breakeven occurs around the fourth time period — perhaps that’s the second half of the second year.

More optimistic projections take the hockey stick approach. The folks with hockey stick graphs always show the long arm of the stick going up — this product is just going to take off and sales will go through the ceiling!

More often than these hopeful folks realize, the hockey stick goes the other way.

The underlying theory of sales growth in these charts is unclear — if there is even some kind of well-thought out model underlying them. Often there isn’t one and the creators of these projections are just playing with arithmetic.

An improvement over the simplistic linear or wishful hockey stick “model” is the four-stage product life cycle — launch, growth, maturity, decline. With that in mind, you might get to see sales projections that follow this pattern.

It’s a more nuanced, maybe even reasonable, basis for estimations in spreadsheets. But aside from launch, these phases are not easy to identify in real-time. Moreover, these phases are so general as to be generally useless in practice. And that is because this kind of curve has not been tied into what is known about patterns of adoption of new products or services, especially technology products.

There are better ways of thinking about sales projections for new products from startups or established companies. These better approaches are not original with me nor are they new, although they do not seem to have the popularity that you would expect.

Here then is a picture of the general pattern of adoption of technology products — what has long been called the diffusion of innovation or, with the notch between the second and third groups, the chasm that new products need to get across in order to be successful at scale.

While the speed of adoption and the actual distribution of various groups is not always a nice normal distribution, it provides a useful framework to identify when different kinds of people will adopt the new product or service. (This isn’t the place to go into the details about the characteristics of each of the five groups. More about that can be found in such classic books as Diffusion of Innovations by Everett M. Rogers and Crossing the Chasm: Marketing and Selling Disruptive Products to Mainstream Customers by Geoffrey A. Moore.)

There has been a lot written on this S-curve, but curiously very little of it ties the S-curve to the financial demands of a startup or a new product launch in an established company. While many people seem to have a basic understanding of the S-curve for the adoption of new products/services, they don’t link that to the money they need. But the pattern of adoption has direct financial consequences.

What do these financial consequences look like? Consider this picture.

The model can be more granular, but this simple model illustrates the point. The expenses match the various stages of the adoption cycle, including the marketing costs of getting past the chasm. The model also reflects the fact that later adopters usually require more support in their use of products and even greater attention to an easy user interface than at the start. The model also reflects the need to create the product before there is any paying customer at all.

The technology adoption cycle provides the framework, but the actual pattern for a new product can be updated with real-time data that reflects how fast adoption is moving and into which groups.

With this kind of model in hand, the product planner can estimate the pattern of product adoption in the future. In turn, that enables better financial planning to estimate future profits.

It’s worth also saying that in startups, in particular, cash is king. No matter what your profits might be on paper, if there isn’t enough money to pay for essentials — say the salaries of key employees who are writing your software — then your startup is in deep trouble. In startups, misjudgments about the money coming into the business versus the money going out of it can be fatal. So startups are especially vulnerable to inaccurate sales projections.

Whether it’s a new product in a big company or a startup, even a sales projection (and updates) based on an understanding of the adoption of innovations does not necessarily guarantee a big success. But at the least this kind of smart sales projection will help avoid a nasty surprise that leads to a cash crunch.

November 11, 2020

The Limits To Being Different

Product differentiation is often described as the key to business success. Companies are told that unless they really stand out from the crowd, their products or services will become “commoditized” — an undesirable position in the marketplace that results in little or no profit. This has been well-established guideline in the world of technology startups and even new technology-based product development in existing companies.

And that guidance is mostly right. Distinguishing your products from the crowd of competitors often results in greater than average profits. Consider Apple, with less market share than Android, but lots more profit than its smart phone competitors.

Of course, how to go about this is not so simple. One of the best and most inspiring books about how to differentiate — how to be really different — is Harvard Business School Professor Youngme Moon’s book, Different: Escaping the Competitive Herd — Standing Out In A World Where Conformity Reigns But Exceptions Rule.

These quotes summarize her forceful advice:

“What does it mean to be really different? Different in a way that makes a difference. It could mean doing the opposite of what everyone else is doing — going small when everyone else is going big…

“You could even say that breakaway brands revel in our stereotypes, since they make their living turning them upside down…

“These brands are the antithesis of well-behaved, and their mutiny is directed squarely at the category assumptions we bring to the table. And sometimes the transgression is more than a touch provocative; it’s a bit twisted as well. …

“What a breakaway positioning strategy offers is the opportunity to achieve a kind of differentiation that is sustainable over the long term. … it has no competitors; it remains sui generis.”

This advice applies not only to business, but can also apply to politics. That’s why I wrote a post four years ago called “The Breakaway Brand Of 2016” about the 2016 US Presidential election. Although I doubt that he read her book and his approach certainly didn’t please Professor Moon, Trump seemed to have been using it as his playbook for the 2016 election. His was the perfect exemplar of a breakaway brand in politics.

Now the 2020 Election also showed the limits of this approach. In a two-way election in the US, you need a majority (putting aside the Electoral College, for the moment).

It is also often the case that being different means you won’t get a majority, as both Apple and Trump have found out. For Apple, that’s not a problem. For Trump, it meant he lost the election.

While he did receive many votes, the limits of breaking too far away in politics was well stated by the most successful politician in American history, Franklin Roosevelt: “It is a terrible thing to look over your shoulder when you are trying to lead — and find no one there.”

The limits of extreme differentiation are clear enough in electoral contests. But the election result also reminded me that there are limits to being different in business too. I’m especially thinking of most established technology-based, multi-sided platform businesses (like Amazon) and other businesses that depend on direct network effects (like Facebook).

These businesses also need to have a majority (or even more) of the market. That’s because their value to customers depends a lot on network effects. Being too different for most people will mean you do not end up getting the majority of people as customers.

So, differentiating — even creating breakaway brands — is certainly good advice in general. But like any advice, it is not always appropriate. And the art of leadership is knowing when not to follow generally good advice and take a different road — even a different road about being different.

November 4, 2020

Lessons From Online Higher Ed In A COVID-Infused World

I don’t think I have ever written about my teaching duties before. But circumstances change, so here goes.

I have been teaching online since before COVID forced most classes online. Each semester I have an online class I try to experiment and improve.

But the COVID pandemic has forced an extra dose of creativity and a re-thinking of ideas – some new and some old – about education. Here I want to share with other educators some of what I have learned in the process. I’ll keep it general as I hope it will contribute to a discussion about how education will occur going forward.

Flipped Classroom

The flipped classroom is not a new idea. But since long lectures in Zoom taxes almost everyone’s powers of concentration, we made the move to a completely flipped classroom for the completely online courses that are the norm now.

What used to a live (synchronous) class that combined a lecture and some student interaction has become a workshop this semester.

The Overview “Lecture”

The lecture material, really an overview of the week’s topic, is now a recorded video that students watch before they go to the live class. This can be anywhere from 15 minutes to an hour depending upon where we are in the course and the topic.

As before, I tend to use video of other speakers to break up the experience so that the students don’t just watch me or my slides. This also lets the students see that some of the ideas they are being taught come from other human beings, not just textbooks – and they can see those other folks, in all their glory and with their tics, quirks, etc. Video is also useful to practice the old adage that it is better to let someone see the story than to relate it to them.

Making The Lecture More Interactive

Because the “lecture” is a recorded video, we can lose the opportunity that students have in synchronous classrooms to ask questions, make comments and contribute to each other’s knowledge. After creating a video in PowerPoint, we don’t just post the video online. Instead, we use VoiceThread which enables more interaction. Students can insert comments, questions, replies of any kind – using text, voice or video. My students have generally stuck to text. Then the faculty and other students can reply.

It’s not quite the same thing as a lecture in a live, synchronous classroom, but it comes close enough. In the first three of these videos, we have averaged about 50 comments each. That is a good level of engagement, in fact much more than was the case in the face-to-face classroom equivalents of these lectures.

Although VoiceThread integrates reasonably well with the learning management system we use – Canvas – it has its limits for this purpose. We can set up an assignment that requires students to watch the whole video, but VoiceThread only seems to enable this to happen if the students look at all the comments that have been inserted into the video. From the perspective of increasing their learning, that’s not such a bad idea, but it would be nice to require them just to see the video. Apparently, that feature is coming sometime in the future.

And Zoom, Of Course

Like many others, we use Zoom for the synchronous class sessions, which are workshops in our case. A typical session starts off with a review of any issues that arose in student assignments in general. Then we turn to the draft of an assignment the students worked on before class. That assignment is usually the completion of an analysis in a workbook which is relevant to the topic of the week.

By now, most people are familiar with Zoom so there is a little learning curve. And, as software goes, it is stable. Even when it runs into a problem, it will reboot itself and pick up in the meeting where it left off.

From the teacher’s view, there are at least two benefits in comparison with the traditional classroom. First, you can more closely scan the faces of students to see if they are engaged.

Second, it is easy for students to show their work to the whole class by sharing their screen. In traditional face-to-face classrooms, it would take a couple of minutes for a student to get up and make the transfer to some device that everyone could see – and in that process the momentum of the discussion would be broken. Now, it happens in a second.

With this ability, we ask two or three students to show their work to the whole class in Zoom. Then both the faculty and other students ask questions and provide feedback on that work. This not only helps the students who are getting this feedback, but it helps other students to realize what they too might have missed or need to do.

Breakout Groups

Then the students are put into small breakout groups where they present their work to each other. This is very useful especially for the rest of the students who weren’t lucky enough to be selected for the class-wide presentations.

We use Zoom breakout groups with random assignment. When students are only paired, there can be a lot of breakout groups. Zoom can handle this number.

However, it has its limitations which in part reflect the challenge the company faces in addressing its diverse markets. In our situation, we have more than faculty member and want each to drop in and out of these break out groups to see how things are going. We finally figured out that we need to make them co-hosts before the break out, but it still isn’t the smoothest process.

We had hoped to use BigBlueButton (BBB) for breakout groups. BBB is video software specifically designed for education. Frankly it wasn’t great a few years ago, but it has been much improved recently. It looked like a better way than Zoom for us to do class breakout group and its user interface and features were better. But unfortunately, BBB has a hard-coded maximum number of breakout groups, which is 8 – too little for our purposes.

Music

We all face that period on Zoom before class starts and the students are straggling in. (This behavior seems to be a carry over from physical face-to-face classrooms. 😉) What do you do to get the attention of students, maybe even encourage their on-time attendance?

One of my colleagues suggested using music in the three minutes or so before class starts. She had in mind some strong, percussive music to wake up the students.

That seemed like an idea worth trying. But I didn’t want just any music. I thought it might be useful to have a song that was appropriate to the topic of the class. And a couple of months ago, I spent more time than I should have searching for just the right percussive, but appropriately themed, music to use. It was a mix, although mostly classic rock.

And it worked! Students show up early chatting with each other about what the song might be and about the song when they hear it. In my last class, I even got a request to set up a Spotify playlist of these songs.

The Results So Far

Overall, the results so far have been very encouraging – better than expected and in many ways better than traditional classrooms. Students seem to grasp the subject matter better, which is the primary aim of course.

But they are also engaged much more. Attendance has been near perfect.

Another measure tells the story better. The online class is officially 90 minutes long, ending at lunchtime on a Saturday. At the official end, I tell the students that they are under no obligation to stay longer. Yet, in the three classes we’ve had so far, a majority of the students stay for more than a half hour to an hour more. Several stay on in Zoom longer than that – some for two hours (when I told them I had to shut it down).

Your experience may vary, since each class and cohort of students is different. These were about sixty students in a master’s degree professional program at Columbia University. But before you jump too quickly to the conclusion that these lessons aren’t relevant to your students, you might want at least to try them.

Do your own experiments and contribute your own observations to this discussion about teaching in a COVID-infused online world – and the world that will be changed after COVID is controlled. After all, it is not just the people in front of you who are students, but all of us are lifelong learners.