Monthly Archives

August 2015

Correction: CB Insights’ “private company health score”

By | Startup general interest | One Comment

CB Insights have released a product called Mosaic which gives private companies a score which predicts their likely future success. The score is derived from lots of data they pull from structured and unstructured sources on the web and then process through a heavy statistical analysis. They pulled data from as many sources as possible looked to see which pieces of it have correlated with startup success in the past and assumed that companies scoring highly on those elements will succeed in the future.

The score has three components, momentum, market, and money. That seems reasonable, although doesn’t give as much prominence to team and product as most VCs do when they talk about startups. I guess CB Insights would counter that team and product drive the momentum score.

There’s much more detail on their site here.

I’m interested for three reasons:

  1. Forward Partners might be able to use Mosaic score as part of our assessment of companies
  2. VCs might start using the Mosaic score as part of their evaluation when they look at our companies
  3. Forward Partners might be able to use the market component of the score to identify new market opportunities as they open up and use that for proactive deal sourcing

Having got that straight in my mind it’s clear we should sign up for a free trial and see what we can do.

The analysis looks thorough and I’m hopeful something will come out of it. Right now I can think of two reasons it might not work for us. Most importantly, I think the analysis is based on correlation not causation, and the score will stop being useful if companies deliberately manipulate some of the inputs (i.e. game the system). For example, the momentum score is partly based on the number of job openings at a company, which is easily manipulated. The second question I have is whether it will work well for the very young companies that we invest in, many of which are too small to have many momentum signals, are in markets that aren’t established enough to have sufficient volume of activity for CB Insights to track, and are too fledgling to have any financial strength.

Update: Earlier versions of this post erroneously referred to Mattermark instead of CB Insights.

Negotiating styles

By | Startup general interest | No Comments

My friend Jason Tavaria posted this link to charts about national negotiating styles to an email group I’m a member of (apologies to other members of the ICE Group reading this for the repetition).

This is how we English negotiate:


I certainly recognise that in myself and my compatriots. Perhaps more interestingly it also helps make sense of where I’m different. I’m very direct by nature and have a strong preference for clarity and speed which leads me to try and push negotiations through the two middle stages as fast as possible. I want to go straight from “reasonable proposal” to “re-packaging” by staying calm and using humour as my tool and can lose patience in the middle stages when people I’m negotiating with are using coded speech and stalling.

And here’s how people in the US negotiate:


The “cards on the table” approach fits well with my style, but no matter how many times I go through fighting in the middle of a negotiation always leaves me wondering if the other person is a … not very nice person. That feeling can go away quite quickly, but it’s definitely there. Knowing that feeling is coming from my British dislike for confrontation definitely helps.

In any negotiation (or interaction in general) knowing yourself and knowing the other side helps find the white space and get a result. These charts are a good tool.


Solutions for closing the gap between browsing and buying on mobile

By | Ecommerce | 18 Comments

Yesterday I wrote about the yawning gap between the 60% of retail browsing and the 15% of purchases that occur on mobile. Michael Haynes commented:

one of the biggest pain points in mcommerce currently is how difficult it is to fill out all the forms – especially if checking out on a new website and factoring in that many sites require registration. It becomes a nightmare that most people will just leave and checkout on a desktop

That makes sense to me. The pain in filling out forms is the biggest reason I sometimes move to my laptop to complete purchases, and bear in mind I’m more patient than many other users because I’m professionally curious about mCommerce. When I do abandon a purchase on mobile it’s either because filling out the form takes too long or because it’s buggy on mobile.

Conversely the beauty of apps like Uber and Amazon is that they have my data already and I can check out with one click.

The key to getting people to purchase on their phones, then, is to take care of the site registration and form filling. There are two broad types of solution getting talked about at the moment:

  • AI based assistants based in the mobile OS – Google Now and Siri lead the pack
  • Messaging clients – Facebook and WeChat are out front here, but services like Telegram and Snapchat are also interesting

I think these all have stated ambitions to enable commerce from a chat style interface, but aren’t doing it yet in volume. Ultimately they will store your personal information and credit card data for you and supply it to ecommerce companies when you want to buy something.

The big question is how discovery will work. If I want to buy flowers on Whatsapp what options will I get? Best case for me is I search and get a full list of providers who have integrated with an open API, appropriately ranked. Worst case is I get to choose between a small number of companies that Facebook has chosen to partner with.

Search currently happens in the browser of course. An alternative solution would be for my data to be stored in the browser and made available to automatically fill out forms. That would keep the open-ness of the web, which would be great for discovering new services. Nobody’s talking about this idea though, at least not that I’ve heard.

Mobile has 60% of retail browsing and 15% of purchases

By | Ecommerce | 7 Comments

I swore out loud earlier today when I read this statistic.

60 percent of retail browsing happens on mobile devices, those devices only account for 15 percent of dollars spent

That’s a massive opportunity, right there.

And what’s interesting is that it’s still with us. Mobile first has been here for years now.

We have good mCommerce models now for regular purchases – Uber and Amazon are two great examples – but for more occasional purchases apps don’t make sense and mobile browsing in its current form clearly isn’t working.

One approach, as taken by our portfolio company Stylect, is to combine entertainment/content with commerce to build an app that people visit daily and purchase from occasionally, but I don’t think this will work for every vertical and what this headline statistic tells us is that there’s a big opportunity to innovate and a large prize for the company that gets it right.


Making sense of the bubble talk and the impact on startups

By | Startup general interest, Uncategorized, Venture Capital | No Comments

There’s a lot of contradictory advice out there at the moment. On the one hand you have the ‘entrepreneurs should just do their thing and not pay attention to the markets’ folk and then on the other hand there are plenty of observers saying that a bubble has burst.

Many people I respect are in the former camp. Tomas Tunguz said it clearest with his recent post Why the bubble question doesn’t matter which lists the things good companies do and points out that they are the same in bull markets and bear markets. I’ve read posts from Brad Feld in the past saying he doesn’t pay attention to bubble talk and in a post earlier this week Fred Wilson quoted someone else quoting him saying “Markets come and go. Good businesses don’t.” (although he did also point out that if companies need to raise money then the capital markets can affect them).

I have sympathy with this view. Startups and venture funds run for 5-10+ years, are likely to see a recession at some point in their lives (maybe two) and hence need to be able to survive and prosper in both recessionary and growth environments. Moreover, predicting when crashes and recessions will happen is nigh on impossible so trying to manage according to where we are in the cycle is a fools game.

But at same time market crashes changes things for startups. I saw that in 2000 and then again in 2008. When the macro economic climate is tough less money flows into venture funds and startups, so fewer deals get done, valuations are lower and more companies fail. On top that everyone is nervous and deals take longer to complete. Making things worse still, consumers and enterprises have less money to spend and startups find it harder to grow revenues.

It takes time for the impact of crashes to be fully felt in the startup market though. I remember this most clearly from 2000 when I was in a fund that was investing heavily pre and post crash and the VC adjustment took 8-9 months. I think the reaction is slow because VC funds aren’t directly linked to the stock market, VC deal cycles are long, because LPs don’t want VCs to try and time markets and because VCs have staffed up to deliver multi-year investment plans. After a while though VCs find themselves spending more time with portfolio companies struggling with the new environment and the amount of new money in the market drops, and these forces combine to stretch out deal times, reduce the number of deals done and reduce valuations.

Mark Suster set out a number of the dynamics at play in his post Making Sense of the Stock Market Drops in Relation to Venture Financing

Pulling it all together I think the difference between the camps is that the ‘pay no attention to the markets’ folk are talking about best practice startup management in general whilst Suster and others are talking about the impact of crashes in the here and now.


I have no idea if the stock markets will continue to go down or recover but it’s pretty clear to me (and probably to you too by now) that if things don’t get better we will get the negative impacts described above, and if they do recover late stage VC markets will continue to get frothier and that will eventually trickle down to Series A and seed.


I think the best outcome is that the bear market continues long enough to take the heat out of late stage venture but isn’t severe enough to create a rout.

Until we find out founders should follow the old adage ‘hope for the best, but plan for the worst’, prepare themselves for longer fundraising cycles, and think seriously about taking any offers of cash that are on the table, even if the valuation is lower than hoped for. (And, in case you’re wondering we don’t have any low valuation termsheets out there at the moment. Our valuations have remarkably consistent over the two year life of Forward Partners.)

Is the path of technological evolution inevitable?

By | Startup general interest | No Comments

I read an article this week which essentially said the current ‘is it good/is it bad’ debate amongst politicians on the future of the sharing/on-demand economy is as futile as 19th century politicians holding a yes-no vote on industrialisation. They were making the point that the sharing economy cat is already out of the bag and the debate should be focused on how to shape it to bring the most benefit to society, not whether we should somehow try and stop it.

This goes right to the heart of whether technological developments are inevitable. Many commentators, myself included, believe that certain things are going to happen – kids will use more social media, more and more devices will be connected to the internet, ecommerce penetration will grow substantially from where we are today etc. etc. Many others are uncomfortable with this view, believing that we have invented technology and should be able to control it.

I’m currently reading Kevin Kelly’s What Technology Wants which sets out a framework for thinking about this question.

The TLDR is that the broad direction of technological development is pre-ordained, but we have control over the precise nature of how it unfolds.

He makes a helpful analogy with our lives as humans. When each of us is born we are given certain constraints within which we have no choice but to operate. Our genes determine that we have to eat, drink and sleep, have a strong predilection to reproduce, and will have a fairly predictable lifespan. After that the context in which we grow up has a great influence – whether we are we born in an age of physcial or mental labour has a massive impact on our lives, as does the extent of parental pressure to work hard, or follow a particular path, or the strength of the economy and range of employment options when we enter the workforce. Finally the decisions we make also play a big part. Some individuals find the energy to dig deep and overcome genetic limitations, others don’t. Some individuals go with the flow of society, whilst others rebel.

These decisions that we make are doubtless very important. They determine who we are and how we are perceived, and they are what is remembered about us as individuals. But they can only go a small way to transcending our genetics and context. No matter how hard she worked no 5th century woman could fly to Australia. Equally it would be very hard for an aspiring 21st century politician to make a big impact without embracing television and the internet.

Similarly, the direction of technological evolution has three determinants:

  • Structural – there is an inevitable trajectory towards greater organisation of information – first the printing press, then the telephone, then television, then computers, then the internet, then social networks, and so it will continue. Once each of these was invented it was only a matter of time before the next one came along. At a more detailed level there are paths of development which have their own inexorable logic – e.g. two wheeled cart, to horse drawn cart, to motorised vehicle, to self-driving car.
  • Historical – the historical context influences the pace, direction and application of technological development. We wouldn’t have had jet aeroplanes or atomic bombs in the 1940s if it wasn’t for the Second World War.
  • Intentional – at any given moment the citizens of society focus the use of technology in one direction or another, determining for example, whether it is applied for good or evil or whether social justice and harmony is prioritised over wealth creation (the sharing economy debate).

If this framework is correct then as individuals and as policymakers we should accept that technology will continue to develop, that there will be good and bad in that, and that the best thing we can do is try to shape it and bend it to maximise the good and minimise the bad. On the positive side there is a great deal of good to be had, as seen by massive reductions in global poverty and child mortality in recent decades, but on the negative side that does mean the bad can’t be eliminated and whilst we can minimise unpleasant new developments like cyber bullying and even grooming, we can’t eliminate them.

Make sure you have discipline and flair in your startup

By | Startup general interest | 6 Comments




Running a startup requires flair and discipline. Flair gives you the big vision, a great story, the ability to close big deals, to woo employees and to get great press. Discipline enables you to get the small stuff right so the business scales well and is optimised in areas like marketing and supply chain.

Fans of HBO show Silicon Valley which follows the story of a company called Pied Piper might recognise these skills in characters Ehrlich (flair) and Jared (discipline). I’m only up to episode 5 of the first series (late convert) but already Erlich has saved the day with an off-the-cuff speech to a key investor about the company’s vision and Jared has implemented a scrum development process without which they would have missed a key deadline.

Most entrepreneurs have a clear strength in either flair or discipline and but to have both is rare. It’s human nature to value the things we’re strong at and that can lead to entrepreneurs with flair seeing discipline as something that gets in the way, and entrepreneurs who are disciplined viewing showmanship and flair with suspicion. Erlich and Jared don’t get along.

The best companies embrace the need for flair and discipline and manage the inevitable tensions that arise. Founders who do this well hire for the skill they don’t have and let their vision and values determine when they should let themselves be over-ruled.

Advice on changing organisational culture

By | Startup general interest, Uncategorized | 3 Comments

I’ve written a lot in the past about how smart entrepreneurs harness company culture as a tool to drive success. Most of that work has centred around being clear on vision, mission and values and it’s never too early for founders start thinking about these things. Sometimes things go awry though and the culture needs to be changed. That’s a difficult thing to do and I’ve just come across a brilliant 2011 post by Steven Denning which sets out the problem and provides a framework for finding solutions.

If you’ve got time, go read the whole thing. For the attention starved amongst you, what follows is a summary.

Culture change is hard and often fails because culture resists change. Here’s why:

an organization’s culture comprises an interlocking set of goals, roles, processes, values, communications practices, attitudes and assumptions.

The elements fit together as an mutually reinforcing system and combine to prevent any attempt to change it. That’s why single-fix changes, such as the introduction of teams, or Lean, or Agile, or Scrum, or knowledge management, or some new process, may appear to make progress for a while, but eventually the interlocking elements of the organizational culture take over and the change is inexorably drawn back into the existing organizational culture.

But if the culture isn’t working then the company won’t work until it’s fixed, and this framework lays out the tools at a manager’s disposal to create a solution.


The best approach is to start at the top and systematically work down, only using the pure ‘Power Tools’ of coercion, threats, fiat and punishments as a last resort. Common mistakes are to use the ‘Power Tools’ too early and to articulate a new vision without putting in place the management tools to get buy-in and re-enforce the message.

Those mistakes are common, but so easy to make, particularly as an investor. Just writing these sentences is bringing back painful memories of working with CEOs to articulate a new vision, strategy, or direction and then watching as months rolled by and little changed. With the benefit of this diagram it’s clear to me that when things didn’t work it was because I didn’t do enough to make sure the management tools were in place, particularly those designed to ensure top-to-bottom buy-in. That contrasts with companies where we successfully used OKR type structures to get full alignment.

YC follows a well trodden path for investment firms: drifts later stage

By | Venture Capital | 2 Comments

If you read this Techrunch post profiling 50 of the current YC companies you will notice that many of them are up and running with customers and revenues. That marks a shift in the YC investment strategy which used to focus on younger businesses. In the words of an alum from the 2006 cohort:

Companies are joining YC at a much later stage.  When I started YC, most companies wrote their first line of code in the first week in the program.  Today, most new YC companies have been operating for a year or longer and have customers and revenue before starting.  If the companies in my batch applied to YC today, I doubt that any of them would get in.

That relates to a second change he observed:

Companies are much more ambitious now.  In 2006, getting acquihired by Google for a couple million buckets was considered a fabulous outcome and basically the goal of every company.  Today, there are six YC companies worth over a billion dollars, and as a result new startups aim much higher.

I think there are three inter-related reasons for these changes.

  • Now that the YC portfolio is worth over $30bn (a number they put on their homepage) acquihires don’t move the needle. The work of investing and mentoring only feels worth it if it can impact the $30bn figure, and returns of less than around $1bn aren’t significant in that context. Note that these returns are from the value of YC’s stake in the company, not the total valuation. Hence they back companies with more ambition. Much more.
  • Because Sam Altman & co have enjoyed a lot of success good entrepreneurs and investors alike are rushing towards YC in the hope that the magic will rub off on them. This creates a virtuous circle of demand and enables YC to invest in better companies than before. Better often means less risky, and hence more mature companies.
  • Assessing the potential scale of an opportunity is one of the hardest things to do for startups and the earlier you invest the harder it is. To have more confidence that their investments match their new found levels of ambition I imagine the YC partners are drawn towards companies that have more validation of their market size.

As I mention in the headline this is a well trodden path for successful investment firms. The usual VC path is slightly different to YC in that it’s also linked to fund dynamics, but the flow from initial success to larger fund to targeting larger exits has been seen many times over. Perhaps the best example is Apax, from here in the UK who started out as one of the first venture capital firms and over time morphed themselves into a private equity company doing multi-billion dollar deals.

Strategies for seeding marketplaces

By | Startup general interest | 9 Comments

I just read a VersionOne post from May about seeding marketplaces. They identify four strategies (there’s more detail on each in the original post):

  • Identify unique inventory – sellers who don’t otherwise have an online outlet will list on your site (provided it’s easy to do) and you can use their product to drive demand. If you are lucky the sellers will bring some customers with them. Etsy is a good example.
  • Bring inventory from another site – hacking and scraping are common grey area tactics. AirBnB is a good example – see case study.
  • Pay for inventory – I think this only works at the very earliest stages, and even then I’d be careful. Apparently Uber did this in Seattle, paying drivers to sit idle whilst they built demand.
  • Aggregate inventory from other sites e.g. through affiliate programmes – scale comes quickly with this strategy, but adding enough value to become sticky can be challenging.

I would add another, and this is my favourite, and that’s ‘Using demand to acquire supply’. Our portfolio company Lexoo used this strategy, first finding companies that needed a lawyer and then calling up lawyers offering them customers if they register on the site. It’s brutally simple and highly effective. Only works in services marketplaces where customers don’t expect an instant quote.

My other observation is that in most cases one side of the marketplace comes much more easily than the other. On Lexoo supply comes more easily whereas on Appear Here, a marketplace for short term lets on the High Street demand is the easier side. The trick then is to build the easy side to make the marketplace super attractive for the more difficult side.