Author: Craig Lawrence

Home / Articles posted by Craig Lawrence
The Siren Song of Modularity

The Siren Song of Modularity

Product designers and engineers are drawn to the idea of modularity. I am guilty – I find modular product concepts irresistibly compelling.

It seems every few years, a new product or concept comes out that stokes that interest – promising all the benefits of modularity, yet with none of the costs. The most recent high-profile example I’m aware of was PhoneBloks, which eventually became Project Ara within Google.

The concept is quite cool. A fully modular phone, where every element: battery, display, camera, speakers, processor, etc. – is user-removable and replaceable. Enabling you to customize your phone for how you want to use it, personalize it, and easily upgrade to whatever the next cool thing is. Creating the physical equivalent of an “app store” for modules, and opening up a brand new ecosystem for innovation.

This was a fairly radical and novel idea for phones, but modularity is certainly not a novel idea. It shows up all over the place, in different ways.

While I was at IDEO, we worked on the Handspring Visor PDA (not public display of affection – but personal digital assistant, that short-lived product category tech-types bought before smart phones). The Handspring differentiated itself from it’s close cousin, the Palm Pilot, in part by using a modular system called the “Springboard,” which would allow you to plug in all kinds of interesting hardware modules: cameras, extra memory, cellular modems, etc. As technology improved, you would always have access to the latest cool gadget without having to buy a new PDA.

The very non-modular iPhone pretty easily beat the Handspring and other modular mobile devices, relegating them to the kitchen drawer of tech history.

In a different industry, Project Better Place attempted to bring modularity to electric vehicles, by making a swappable battery pack for EVs. So, instead of plugging in and waiting 30 minutes or more to recharge your car, you’d pull into a Better Place swap station, and get a fresh new battery installed by a robot in minutes. As battery technology improved, you would benefit, versus being stuck with that aging, obsolete battery for however long you owned the car.

The fully-integrated, non-replaceable battery of the Tesla Model S, combined with their super charger network pretty easily beat the Better Place vision (although modularity certainly wasn’t the only reason).

Just so you don’t think I’m cherry-picking only product failures to use as examples, there are some great examples where modularity has been successful.

The Personal Computer, in it’s original form, was the ultimate modular product. Swap out graphics boards, memory, power supplies, hard drives, peripherals, as needed. Build your own from a wide range of components and vendors. And, of course, software applications on PCs and other platforms represent the ultimate in product modularity.

I would argue that the PC worked as a modular product because it is really a platform, not a product. It doesn’t do anything on it’s own, without it’s hardware and software modules. And, it’s days as a modular product appear to be numbered, perhaps relegated to a niche of high-performance gamers and hobbyists. The classic desktop PC is giving way to ever more powerful laptops, tablets, and smart phones, most with zero modularity (can’t even replace the battery) – purpose-built, optimized, svelte and sleek, and great at doing what they are designed to do.

Modularity is the go-to for design students and design competitions. It never goes away – concepts for modular housing, modular transportation, modular furniture, modular appliances, modular electronics, etc.. are continually recycled, replayed, reconsidered. Yet, unfortunately, they almost never make it to market.

In most cases (but certainly not all), purpose-built products are going to beat modular products, every time.

Adding modularity to a product has real costs. For physical products, making something removable and replaceable almost always adds size, weight, and cost, and impacts the performance of the product’s core function. There is a reason why Apple does not make the batteries on their products user replaceable – the phone would be thicker and heavier (and would last longer – something consumers might want, but Apple doesn’t). There is a reason why Tesla does not make the batteries on their vehicles user replaceable – the car would be bigger and heavier, and would have more failure modes – there is nothing more tricky to do well than to design a reliable electro-mechanical connector.

And, modularity works best in a diverse, open ecosystem. Those who have tried to build their own closed modular product platforms have typically failed. And getting others to invest in developing modules for your product is not easy.

I’m not suggesting designers stop trying – it is a noble goal. But, beware of the siren song of modularity. Many have tried and failed before you. And, like you, I am still strangely attracted to it…


Tesla, Please Don’t Give Solar a Black Eye

Tesla, Please Don’t Give Solar a Black Eye

I was very excited to see Tesla provide more details on their Solar Roof. A bunch of smart people have already started to dig into the details to try and figure out the true cost, including GTM and PickMySolar.

I have been skeptical of this product – so many have tried and failed. But, the same could be said of EVs, so I’m certainly rooting for Tesla, and will always give them the benefit of the doubt. There seems to be tons of technical risk here, but if they can deliver, even at the prices they are showing on their website which seem to be substantially higher than “regular” solar, then that is going to be a win for everyone. It’s great to see product differentiation in solar – not every consumer wants the same thing, and some people are willing to pay a lot for aesthetics and other less quantitative values. As is true in most every product category.

I’m not worried about the product being a failure. Even if it is, the industry will survive and move on. What I am worried about is a political failure. Specifically, a black eye for solar with the Treasury Department regarding the ITC, which many of us worked so hard to get extended. Is Tesla going to try to claim the Investment Tax Credit on the full cost of the roof? Which, by the analysis linked above, looks like it comes in at about $10 per watt? If they try to do that, then Tesla Solar Roof purchasers will get a $3/W rebate, versus a $1/W rebate for everyone else. For a 6kW system, that’s an extra $12,000. And for producing the same (at least, we hope) amount of solar energy as the sucker who bought the standard solar panels for their asphalt shingle roof. Even if they split out the cost of the non-solar tiles and the solar tiles, and only apply for a tax credit on the solar, it would still be substantially more credit than the rest of us get.

As a taxpayer, I don’t think Tesla or their customers should get a larger tax credit for buying a really cool roof. As someone working in the solar industry, I don’t think they should get more. This is the kind of action that got SolarCity and many other developers in trouble awhile back – marking up their cost basis and capturing a larger tax credit. And I guarantee that the enemies of solar will take this opportunity to paint solar incentives as a give-away to the rich, who don’t need it, at the expense of coal workers.

Let’s hope Tesla steps up and proactively does the right thing here.

The One Day Business Case

The One Day Business Case

As an engineer, I had the typical mild level of disdain for “business people” – marketers, MBAs, people with fuzzy backgrounds that appeared so confident, despite not knowing the basic laws of physics. I took pride in my engineering-ness. I huddled with my fellow engineers, quietly smirking after the business people left the room, telling ourselves their input was less valuable than ours.

Then I somehow ended up at a Venture Capital firm, ostensibly to help them with technical due-diligence on complex clean technology start-ups trying to innovate in solar, energy storage, fuel cells, electric vehicles, biofuels, and even nuclear energy. I found myself listening to hundreds of start-up pitches. I discovered that the “other stuff” in these pitches – market research, strategy, competitive analysis, SWOT, TAM, SAM, ASP, COGS, Opex, Capex – were interesting. And important. And uncertain, but not as “fuzzy” as I once thought. I saw good pitches and awful pitches. And I learned that technical success was only one small ingredient in overall success. And, I wanted to be successful, so I poured myself into learning how these “business people” thought and worked.

Fast forward to today. Realization Ventures gets involved in launching many new products and services at both big companies and startups. And, I’m always surprised at how few of these organizations have built a basic “business case” for their new thing.

A business case is simply a collection of facts and hypotheses that form the rational for why you should do this new thing. How big is the market? how many customers can I get? how much will they pay for it? how much will it cost to produce? how many people and how much equipment will I need? And how will all those things change over time? A business case is much more than just a spreadsheet or document – a good business case provides a testable hypothesis on which to start building a business.

By separating fact from hypothesis, you can develop a set of low-cost “tests” of your business case, to validate that you know what you are doing. You can prototype a business just like you prototype a product – the business case is your design. And it is not a static thing – it should change over time as your hypothesis are validated (or not). It allows business people to utilize the scientific method I was so proud of as an engineer.

Launching a new product, service, or business without a business case is like going hiking without a map. If there are signs on the trail, you can follow them, and hike the same path everyone else has. But, if there aren’t signs (which there usually aren’t), or the signs lead to dead-ends, you are just going to get lost.

Every time Realization Ventures starts working on a new product, we encourage our clients to do what we call the “One Day Business Case”. The idea is to bring all the relevant disciplines into a room for a day – marketing, sales, engineering, operations, etc. – free from distractions, and with all their facts and assumptions at hand, and build a business case as a team.  We not only build the business case, but define the tests we need to validate our hypotheses. It’s not easy, and often requires a neutral facilitator. It’s messy but fun. It allows everyone to think about the big picture, and how their piece interacts with the whole. It is a great team building exercise. People leave feeling aligned.

If you are working on something new – I encourage you build a business case. And don’t send an analyst off in a corner to build it – consider the One Day Business Case.  If you want any help, Realization Ventures is here for you.


How an Electric Vehicle Re-Wired my Brain

How an Electric Vehicle Re-Wired my Brain

There are technologies that come along every so often and completely change your experience with a product. And change your baseline for what is normal, despite a mental resiliency built from years of habit and repetitive experience.

One example of this was Tivo, which transformed the way we consume media, with the ability to time-shift content seamlessly and pause live TV. I knew my reality had been altered by Tivo when, after having the service for about a month, I found myself unconsciously trying to pause my car radio and realizing that, unfortunately, not everything worked like that. But, it should.  Another technology that had a similar effect on me is the “Smart Key” technology on my Toyota Prius. No need to press a clicker to unlock your door – as long as your key-fob is on you, it just unlocks when you grab the handle. I recognized the effect this had on me when I walked up to the front door of my house one day and expected it to unlock when I touched the knob. And again realized that, unfortunately, not everything worked like that. But, it should. These transformative technologies “re-wire” your brain, and change what you accept as normal behavior for a product interaction.

My EV has re-wired my brain. And, I’m convinced it will happen to everyone.

A year ago, I purchased a Chevy Volt (2nd gen), technically a plug-in hybrid, with a 50-60 mile all-electric range, and a 9 gallon gas tank that gives you about an additional 400 miles of gas-powered range when you run out of juice.

I did have reservations about this purchase –  would I be able to find places to plug it in? Would it actually save me any money? I hedged my bets with the Volt – not going all-in EV, and was comforted by the fact that I could rely on gas in a pinch.

Over the last year, I’ve been to the gas station a total of two times. 95%+ of my miles have been all-electric. And 95%+ of my charging has been done at home or work, when my car is sitting idle and I’m off doing something else. What I imagined being the biggest inconvenience – where can I find a charger? – has been by far the biggest benefit. I don’t think about fueling my car – it’s no longer a conscious action I need to take – I plug it in when I get home, and it’s “full” when I wake up in the morning. It requires minimum effort and has become a part of my fabric.

I experienced my altered baseline the other day on a business trip. I had a gas-powered rental car, and every night when I got back to my hotel, I instinctively wanted to plug it in. Unfortunately, not everything works like that. But, it should. I was doing quite a bit of driving, so had to fill up on my trip. I had a visceral, negative reaction as soon as I pulled into the gas station. It was dirty, ugly, smelly, and I had to go out of my way, waste about 15 minutes of my life, swipe a credit card, punch in my zip code, decide what octane gas I wanted, and of course, it dripped on my foot as I went to put the nozzle back. It just felt wrong. And outdated. And dying.

I’m saving money with my EV – I estimate about $750 a year. But, I am also saving time – I calculate that I used to spend 16-20 hours a year fueling my car. That’s 2 full 8-hour work days wasted. Now I basically spend zero. Multiply that by 200+ million drivers in the US, and we are wasting over 1 million person-years every year at gas stations.

But, more significantly, I can’t stand operating anything other than an EV. It just feels wrong. Like not being able to fast forward through a commercial.

My brain is re-wired for EVs

Lower solar customer acquisition costs with…. microinverters?

Lower solar customer acquisition costs with…. microinverters?

I spent several years at SolarBridge Technologies, an AC Module company (acquired by SunPower in 2014). So, disclosure: I may have some bias. Microinverters and AC Modules generally cost more than the competing solutions – they are a premium product that provides unique benefits. You need to get very good at explaining those benefits to customers in order to close deals. Most advocates talk about better energy harvest due to module-level MPPT, higher system reliability due to no single-point-of-failure, and perhaps reduced design and installation costs due to the simpler architecture. I truly believe all these things are true.  And, the sales pitch seems to be working. Ten years ago, microinverters had a zero % share of the US solar market. Today, Module Level Power Electronics (MLPEs), including DC Optimizers, have well over a 50% share of the US residential market.

One benefit of MLPEs that I stumbled upon, and that nobody seems to talk about, is their ability to reduce your cost of customer acquisition. “Huh?” you say? This connection is not obvious to the casual observer. Here is the kind of feedback I got from customers at SolarBridge that let us know we might be on to something (paraphrased): “We love the AC Modules – we can get more solar up on the roof because I don’t have to worry about string constraints.” Or, “We went back to a house that we couldn’t make work with string inverters due to the broken up roof faces, and our designers figured out how to make it work with AC Modules.” I heard a lot of that – it made me feel good. I also got a lot of negative feedback when our stuff didn’t work well, but that’s a topic for another blog.

We started thinking about this. Getting a solar customer is expensive. The latest data I’ve seen is that it’s anywhere between $0.20 to $0.85/W to get a solar customer. $1,200 to $5,100 for a typical 6 kW system. It is a big target for the solar industry, as it is one of the most expensive line items in the solar cost stack. There are two fundamental ways to reduce customer acquisition costs ($/W):  spend less money to acquire the customer (lower the numerator), or sell more Watts to your customers (increase the denominator). MLPEs do both. Really.

  1. MLPEs enable more homes to go solar. Say you are running a marketing campaign. You spend money to reach customers – flyers, door-to-door, advertising, whatever. Some fraction of those customers that respond are going to drop out of your sales funnel because their roof is not suitable: shading, odd angles and roof sizes, complexity. MLPEs enable more of these roofs to go solar, increasing your conversion rate. For the same amount of marketing dollars, you will get more customers, reducing the $ per customer cost. Lowering the numerator.
  2. MLPEs enable you to put more panels on a roof. String inverters can be tricky. Strings need to be a minimum size. All the modules in a string need to be facing the same direction. This places constraints on how you can design a system. Microinverters, and to some extent DC Optimizers, eliminate those constraints. You can put any number of panels, anywhere you want. This lets you get more solar on a given roof. More Watts per customer. Raising the denominator.

All of this can be difficult to quantify – but, we tried to at SolarBridge. We did this by surveying 32 of our US installation customers across 16 states. What we found was that 25% of the sites they installed would not have penciled without microinverters. And that on average, they could fit 7% more modules on their AC Module sites vs. their string inverter sites.

The net result of this study showed a 22% potential reduction in customer acquisition costs. $/W.

Look, I know we weren’t exactly an unbiased third party at SolarBridge. I don’t know if our study met the standards of a peer-reviewed scientific journal. But, I know there is something there. Because I heard it from my customers.


Where are the big brands in solar?

Where are the big brands in solar?

In 2006, I was working at design firm IDEO on a research project for newly-IPO’d First Solar, who was interested in getting into the residential solar market. We talked to early adopters, tech luddites, installers, and everyone in between, to understand how they felt about home solar. Few people understood, and fewer cared – it was ugly, confusing, and niche. The market was measured in thousands of homes – it was by any definition, a cottage industry.  First Solar went on to invest in SolarCity, but never really participated in the sector.

Fast forward a decade:  In 2016, the industry installed nearly 3 gigawatts of residential solar in the US, many hundreds of thousands of homes, representing nearly $10 billion in market value. Most of these systems were built by companies that did not exist in 2006. The industry has grown up quite a bit, and today solar is one of the fastest growing parts of the economy. Recognizable brands are starting to emerge, financing has become commonplace, marketing is more sophisticated, and as in every other industry, software is eating the world.

So, why are there no big consumer brands in residential solar? With $10B up for grabs, the commodification of hardware, and prices down by at least a factor of three since 2006, where are the home services providers and product companies? AT&T? Comcast? Google? Apple? Amazon? Anyone? Even the consumer companies that produce solar panels – LG, Panasonic, and Sharp — have not really gotten “in” to solar. Their efforts are isolated manufacturing businesses within global conglomerates, with little ties to their strong consumer brands.

There are plenty of good reasons for the leading consumer brands to be interested in residential solar.

  1. It’s big
  2. It’s profitable (if you are not in the commodity business)
  3. They have an unfair advantage that today’s biggest solar companies do not: customers.

These companies have already acquired millions of customers who are buying things, consuming content, and signing up for long-term services. They have deep data on those customers, have been in their homes and wallets for a long time. They have effective, nationwide sales, marketing, logistics, service, and operations teams. They could potentially slash the biggest remaining cost in residential solar – customer acquisition. And, today’s solar companies represent an emerging threat to them – everyone wants to own the home.

It’s not easy… Some have tried – sort of. Comcast has played in energy retailing, and has a smart home energy offering. And they have partnered with SunRun to deliver solar to their customers, in what appears to be straight lead-generation. Vivint has a thriving solar business, but it is decoupled from their very successful and larger alarm business (I am a Vivint alarm customer, and have never been approached by a Vivint Solar salesperson!). AT&T has partnered with SunPower and DirectTV with SolarCity (again, lead-gen). Google has SunRoof. I can buy solar panels on Amazon.

Many would argue that home solar is not conducive to a national strategy, and is destined to be a regional, long-tail industry. I believe that is true, as long as solar is delivered to homeowners as a construction project, and not a product. But, solar doesn’t have to be a project. And, the battle is ultimately going to come down to customer acquisition.

The real opportunity for the big brands lies in an integrated offering – another line-item on my bill, a quadruple- or quintuple-play. But, this time, an offer that saves, rather than costs me money.

There are two significant technology trends that have now made it possible for these companies to participate at scale, if they choose to:

The “productization” of solar – Microinverters and AC Modules have nearly succeeded in turning solar from a project into a productmaking it possible to standardize offerings, and safe and easy for non-specialists to design and install a system. SunPower has moved much of their residential business to AC Modules because of these benefits. This is really the only way to scale the industry from hundreds of thousands of homes a year, to many millions.

Software-enabled solar – the tools to do remote evaluation, design, financing, and closing of solar deals have never been better. A decade ago, the state-of-the-art was an elaborate Excel spreadsheet. Today, you can go end-to-end in the cloud with offerings from Aurora Solar, Enact, Sighten, Helioscope and others. Technology from companies like Google and Geostellar avoid the need to get on the roof to quote. And, drones are coming.

So, the market is big, the timing is right, and the technology is there.

But, I’m not seeing much movement. And, I’d love to know why.




Welcome to Realization Ventures! We are a consulting firm built to bring new ideas to life. We help companies, big and small, to visualize their future, and design and launch new products, services, and businesses. We work across several industries, including energy, consumer and industrial products, and Internet-of-Things. We are builders – we don’t want to deliver PowerPoint, but rather stand-up new products and business. We are risk takers – willing to tie our success to yours. And, we are flexible – engagements can be anywhere from one hour to many years.

If any of this sounds interesting, please drop us a note! We’d love to hear from you.

Craig Lawrence, Managing Director