As a keen observer of startups over the last 17 years, one of the most remarkable and welcome developments has been the application of scientific method to building startups. In 1999 when I started in venture capital there were no blogs and very few business books that were useful for entrepreneurs. All founders could do was accumulate wise advisors and rely on their wits and instinct.

If I was to pick a watershed moment in the emergence of ‘entrepreneurship as a science’ it would be the publication of Steve Blank’s Four Steps to the Epiphany in 2005. It’s not the easiest read, but for the first time founders had a playbook they could follow. However, it was also around that time that Brad Feld, Fred Wilson and a number of other wise souls started blogging and startup best practices started to be widely shared.

There were two great things about that. Firstly sharing leads to discussion and discussion leads to iteration, making everybody involved smarter. Thus it was that Eric Ries both extended Blank’s work and made it more accessible with the publication of The Lean Startup in 2011. Secondly, people outside of Silicon Valley were able to join in the conversation and get smarter to a much greater extent than they ever had been before which was a massive boon to other startup ecosystems around the world, including London.

Here at Forward Partners we have worked hard to contribute to this development by publishing The Path Forward – a playbook and set of practical guides for founders in their first year or two.

All this work has, I think, made it easier for founders to climb the learning curve and become masters at running their companies. It’s easier to know about and avoid common pitfalls (e.g. assuming you know what customers think) and to pick up tactics and best practices (e.g. OKRs for managing objectives). Of course, that doesn’t mean it’s now easy to be founder, far from it, but it is easier than it was.

However, building a startup can never be reduced to pure science. Some magic, art and wit is always required. I was talking to the chairman of one of our companies a year or so back (I won’t name him for reasons that are about to become obvious) as he was helping them through a rebuild of their product. The founder is a disciplined practitioner of lean startup principles who had achieved good growth through lots of experimentation and optimisation, but they had got stuck. They had hit a local maxima. The chairman explained how they had over indexed on startup science and ended up with a product that was boring. They needed more soul.

This story has a happy ending; they rebuilt the product and are now growing fast once again, but it is a reminder that there needs to be a balance between the disciplined application of startup best practice and inspiration.

I’m writing this today because whilst reading Are Liberals on the Wrong Side of History in The New Yorker I was struck by the similarity between the recent evolution in startup thinking and the way The Enlightenment impacted western thought in the eighteenth century. I don’t have the deepest grasp of the history of philosophy, but it was during The Enlightenment that thinkers like Descartes, David Hume, Adam Smith, and Immanuel Kant had the great rationalist vs empiricist debate which developed the concept of the scientific method, introduced the idea that everything might be explainable through thought and rules, and then hotly debated the limitations of that approach to understanding the world.

As The New Yorker points out, you can, in fact,  trace this debate back to the ancient Greeks with Plato on one side and Aristotle on the other, so the rationalist vs empiricist debate has actually been running for millennia.

When I was an under graduate studying social science in the 1990s I had a good run synthesising the work of the leading thinkers of the time across sociology, political science, social psychology and social anthropology. It worked for me then and I find myself repeating the pattern here.  When there is a significant change in society then the pendulum almost always swings too far, whilst what we really need is to find the right balance. During the great debates of The Enlightenment in a sense both sides were right. It is beyond doubt that rationalist thought and the scientific method brought great advances to our understanding of the world and many great things flowed from that, including the liberal-capitalist system which has given us unprecedented individual freedom and prosperity. However, there are still many things that we don’t understand from first principles where all we can do is treat them like a black box developing predictions for what will happen next based on what we’ve seen in the past without understanding the underlying workings – the human brain is one example, and the workings of the economy being another (hence our difficulty understanding the impact of Brexit).

Returning to startups (and this is a bit of a stretch, but bear with me) – Steve Blank and Eric Ries can be likened to Descartes and other early enlightenment thinkers from the rationalist camp who achieved great advances by using scientific method to shine light into areas that had previously relied upon intuition and rules of thumb. The next step is to balance that thinking with the an approach that can be likened to the work of David Hume who pushed back on the rationalists noting that great insights can also be had by drawing on our experiences.

Throughout his career Steve Jobs famously eschewed market research and relied on his intuition to build amazing products. That’s an extreme position which worked for him, but doesn’t work for most of. The balance I’m talking about cultivates that sense of intuition but then finds ways to quickly and cheaply test the resulting ideas with customers. Now that we are in an era where our basic needs are sated MVPs need to be increasingly sophisticated before customers will engage. That means more investment in development before ideas can be tested than was the case ten years ago, increasing the cost of failure (hopefully not too much) and thus making it more important that only good ideas are tested (again, hopefully not too much). Hence the point of balance is shifting. At the margin the value of good intuition is increasing and the value of disciplined application of lean startup principles is decreasing.

The pendulum is starting to swing back the other way.

Unless you’ve been hiding under a rock, you will have noticed there’s a lot of heat around AI as an investment theme right now. Octopus’s recent announcement of a £120m dedicated AI fund is one of many recent events I could cite as evidence.

In that same announcement Octopus mention that they have had three AI exits (Swiftkey, Magic Pony and Evi) so this is not a new investment trend.

It is, however, a trend that is changing. Up until this point AI exits have largely been driven by a desire to acquire talent. Even Deep Mind’s $400m sale to Google in 2014 is, I think, best understood as an acqui-hire.

Going forward two things will be different. Firstly, universities have responded to the demand for AI PhDs. Hence talent will be less scarce going forward and acqui-hires will be less necessary.

Second, and perhaps more interesting, is that it’s becoming much easier and much cheaper to build AI driven products and we are seeing an explosion in the number of AI startups with a clear path to delivering value to their customers and making profits. There were, of course, numerous companies in the previous generation of AI startups that were on this path, just nothing like as many as we are seeing now and expect to see in the years ahead.

AI startups are becoming cheaper and easier to build, because many of the underlying technologies are now mature enough to apply predictably, and because of the declining cost of cloud computing – including many AI as a service products on AWS and Google Cloud.

I liken this development to the time when cloud computing first emerged around ten years ago. Resources that were previously the preserve of cash rich companies became available to anyone who could pull together a few grand and a thousand flowers bloomed. I think we will see something similar again now.

A couple of times recently I’ve found myself coaching people to stay positive. In both cases they very reasonably pushed back, saying great idea, but they didn’t want to be false and pretend to feel positive when inside they felt anything but. Two conversations about the art of being authentically positive ensued and I’ve been collecting my thoughts on the subject since then.

Let me start by taking a step back. This may be obvious to many of you, but we all like being around positive people. It’s more fun and it helps us keep our own energy up.

Positivity is doubly important in startups where the ups and downs will inevitably lead to periods where we question whether the whole endeavour is worth our time. Happiness is contagious and companies full of positive people climb out of the dark patches more quickly.

However, to really work, the positivity must be authentic. Saying or implying you feel good when you’re really not sure is better than giving into cynicism, but people can tell, and after a while it will chew you up inside.

One trick for staying authentically positive is to avoid dwelling on the big problems and focus on the little wins. When someone asks how you are doing, reflect on something that has gone well recently. If you made minor progress with a major client in the last 24 hours, say so. It’s genuine, and will make you and the person you are talking to feel better than a negative or neutral statement.

Underlying this is a really important point, which is that effective operators respond to feeling down by finding something positive to do. When we were still working out the details of our model here at Forward Partners we had a chap who started to get cynical about key aspects of his role. To his credit he responded by taking ownership of one of our content initiatives. It was a side project for him, but he had success there which kept him positive whilst we sorted out his bigger issues.

Other helpful tricks are getting enough sleep, exercising, eating well, meditating, and – simplest of all – remembering to smile. If you feel good in your body you will have more energy and find it easier to stay positive.

Like happiness, positivity is a function of mindset and behaviour. It can and should be cultivated.

It’s common for VCs to look at the market size for a potential investment from a top down and bottom up perspective. The top down perspective takes market research, often from an analyst firm or investment bank and the bottom up approach works by multiplying the number of customers by their likely spend – more detail in my old blog post here.

What I hadn’t thought of until recently is that it’s also helpful to take a top down and bottom up approach to assessing likely demand for a product.

The top down approach looks at how a startup fits with prevailing big picture trends. At the time of writing AI is the trend of the moment and it’s a good starting point to think that companies which intelligently apply AI techniques can create useful products. Moreover, it’s also true that raising money is easier for companies that are on trend (investors love a herd… or at least most of them do!).

However, the top down approach isn’t sufficient on it’s own. Even though it sometimes seems like companies doing AI for XYZ seem to be raising money almost as easily as companies doing Uber for ABC were a couple of years back, this strategy is unlikely to yield much success for either founders or investors.

To make good investments it’s important to combine the top down approach with a bottom up approach which looks at use cases. If it’s difficult to convincingly explain how someone will use a company’s product, it’s a fair bet that they will find it difficult to get customers. I’m consistently surprised how often entrepreneurs allow themselves to be satisfied with only a vague understanding of why they will make people excited.

When looking from the bottom up, a good first question to ask is ‘what behaviour potential customers are already exhibiting which shows that they will have demand?’ For young software companies a classic answer to this questions is that potential customers are building homegrown versions of the product they intend to build. If our young software company can build a software product that’s better and cheaper than the homegrown version then it’s a fair bet these companies will stop writing their own code and become paying customers.

A second technique is to employ Clayton Christensen’s ‘jobs to be done’ framework which starts from the insight that customers buy things because they have jobs they want to get done. Jobs can vary from the mundane (e.g. cutting the grass) to the exotic (e.g. become my better self) and companies that can articulate a good fit with a job that lots of us have to do or want to do are in with a good shout of selling lots of product. There’s more detail on the jobs to be done framework here.

For infrastructure companies the use cases are often not end user use cases. Rather the use cases are to help other companies build use case for the ultimate end user. For example a company that makes electric motors might sell to a lawnmower manufacturer who’s job to be done is to sell more lawnmowers. The electric motor opportunity can then be evaluated on the basis of whether it will allow the lawnmower manufacturer to help its customers (the end user) with their job of cutting the grass.

As with market size analysis the bottom up approach is harder to do well, but yields much richer insight.

I’ve come across Kahneman’s Peak-End concept before but only just grasped it’s significance. As with much of his work Kahneman is highlighting an area where our minds don’t work rationally. In this case it’s how we remember experiences.

If we were rational we would remember experiences as some kind of average of how they felt at the time, adjusted for their duration. However, it turns out we remember them as a function of two moments – the peak moment (best or worst) and the last moment. Duration and average are less important.

Kahneman makes his point by citing research into how patients undergoing conscious surgery rated their experiences. Their post-surgery rating of the overall experience correlated with the peak moment of pain and how the surgery ended rather than the average of their minute by minute scores for how much pain they were suffering. Indeed, changes in the final moments of their operation dramatically skewed their overall perception of how well it had gone, in both directions.

This difference between how patients experienced their operations and how they remembered them exists with all experiences, both pleasant and unpleasant.

That has big implications for how startups should build products. Customers come back or tell their friends because of how they remember the experience of a product or a service, not because of how they experienced it at the time. Accordingly products should be engineered to deliver moments of delight and happy endings rather than maximum overall utility.

At Forward Partners we always look for those moments of delight, which we call “eyes-light-up moments”. Going forward I will be pushing us to pay equal attention to the closing moments of a customer experience.

I’ve just read Atul Gawande’s Being Mortal, a book lots of people seem to be talking about all of a sudden. It’s a great book, and one I highly recommend. Mostly it is about making better choices for ourselves and our loved ones as we grow old, which is a far cry from startups.

One passage is different though. It’s about courage, which is required in spades if we are to make the most of our old age and if we want to make the most of our startups.

Gawande starts by turning to Plato who wrote “Courage is strength in the knowledge of what is to be feared or hoped” and “Wisdom is prudent strength”.

Which brings me to startups. It takes great courage to found and build a company, to maintain conviction in the face of naysayers, to get back up again when you get knocked down and to persevere when things look helpless. But it is also important to be wise. To know when to carry on or when to change course, or even give up. It was only after Ev and Biz Stone gave up on Odio that we got Twitter.

Gawande goes to on say that at least two kinds of courage are required in ageing and sickness. The courage to seek out the truth of what might happen both on the upside and the downside, and then the courage to act upon it. Again that’s super important at startups – the best founders tirelessly look for ways to expand the upside and ways their companies can go wrong. Other founders put their heads in the sand.

In a further parallel with startups, Gawande notes that ageing and sickness are highly complicated and uncertain, and that it is very hard to build an accurate picture of what’s going on or of the implications of any particular course of action.

It is in this uncertain environment that we must find the courage to act. Very often that choice means deciding which is more important, our hopes or our fears. Do we want the chance of a much better life, or do we simply want to stay alive?

That last question is as relevant in startup boardrooms as it is in the Emergency Room, but it is rare to see the difficult topic of potential permanent decline and failure addressed well enough that the the following course of action gives the best chance of happiness to all concerned. More common is to raise more money if it’s available, keep going with more or less the same plan and only make radical changes when the writing is well and truly on the wall. By this point there is much less cash left, the options for remedial action have become limited and the chances of achieving even a half decent outcome are much reduced.

The starting point for most founders is that they can overcome all the obstacles in their way and overcome what, to many, look like impossible odds. Possessing the self-confidence and resilience to maintain this perspective is an amazing gift that has enabled many entrepreneurs to achieve amazing things and it is something we look for in the founders we back. My closing point in this post though, is that in some situations the right thing to do is to stop trying to achieve the impossible, reduce our level of ambition and start building the best possible future within the constraints that face us.

Contemplating anything other than success is tough for most people involved with startups. Nobody likes a naysayer and fear of being thought of as unambitious or lacking tenacity made me think twice about writing this post. I decided to press ahead because the truth is that only a small fraction of startups achieve unfettered success (we will be happy if one in three of our portfolio are that lucky, and even those will face difficult periods). For the rest finding the courage to face the truth and make difficult decisions early will make them happier in the long run.

This chart, from First Round Capital’s recent post Leslie’s Compass: A Framework for Go-To-Market Strategy is super interesting. It’s first use is for founders to work out whether they should have a sales intensive or marketing intensive go-to-market strategy. That’s the point of the post and the summary is that if your business has the characteristics on the left hand side then your strategy should be marketing intensive and if you’re more like the right hand side you should be sales intensive. If you’re thinking this problem through at all I would highly recommend reading the whole post.

The second use, which they don’t cover, is assessing whether a business idea is likely to be successful. It’s an obvious thing to say, but unless a business can find a successful go-to-market strategy, sales will be limited and it won’t succeed. The power of this framework is that it can expose fundamental challenges to the viability of a plan even when it is only a concept, and then it can suggest ways to address those challenges.

Simple plans are easiest to execute and in this case the simple plans are ones that are either marketing intensive, or sales intensive. Plans that sit somewhere in the middle are ok, but products that have some marketing intensive characteristics and some sales intensive characteristics have an inherent contradiction that if left un-addressed will undermine success.

The most common and obvious contradiction that we see is complicated and high touch products that are inexpensive (or have low margins). Even if the product is a bullseye hit with what the customer needs, it won’t be possible to persuade them of that fact without an expensive sales effort, which won’t be covered by the value of the sale.

Other contradictions to watch out for include B2C : complex products and many customers : low fit, but the most important one is definitely cheap products that require a sales lead approach.

Business plans with contradictions like this aren’t necessarily fatally flawed, they are just more difficult to execute, and that brings us to the third and final use of this framework, which is to inform product strategy. If there is a contradiction then one solution is to resolve it through product innovation – if the contradiction is between low price and complexity/high touch then either find a way to either to take the complexity out or to charge more.

Usually those product innovations will be to enable a more marketing led approach, and to generalise, companies that move product categories from being more sales intensive to being more marketing intensive make promising bets. The shifts don’t have to be big either – convenience is a winning proposition. Examples are legion, but Slack is a great one. Last May they became the fastest company to reach a $2bn valuation in large part because they succeeded in making a product that works with a go-to-market strategy that is close to 100% marketing led. Looked at through this lens, their genius was in taking all the complexity out and enabling low touch adoption.