This is the third of a series of three posts on the topic of competing against firms with established network effects. In this post I explore 3 different strategies, how they work together and at the end I try and show how this could be put into practice in the property market.
In a world of hyper-connectivity very few products live in a silo. A football is enhanced when a vacant football pitch and ten players can be found. A conversation is more exciting when more people can join it. A dress is made better when multiple designers can collaborate, and their views are enriched by the wisdom of the crowd. Insurance is more efficient when powered by a data network and so on. There are infinite examples and all of these require designing for multiple entities: product design techniques, while still useful, have become weak when applied to a single entity. That is why I believe the future of product strategy will increasingly look like platform design.
The second key point I want to make is that while most products and services are best consumed or made through connections which can power network effects – the nature of those network effects can’t be altered. The implication is that a good understanding of the matter will reveal that some NFX are stronger than others and that can only change so much, no matter what the product does. E.g. we don’t care about who drives our taxi, we do care who the babysitter for our children is, we won’t be able to go on holiday more often just because there is a great online service for planning them.
This also means that the concept of escape velocity is partly false – if a company is not positioned to have strong network effects because of the nature of the market it operates in, it will not achieve escape velocity. The attraction of new customers will be too weak and the churn of existing customers too high.
In some cases network effects deliver a winner takes all scenario, (at least initially). This tends to happen when technology enables new patterns of value creation: think of the rise of Google. That’s however seldomly the case – most often the key success factors go beyond network effects. This is why competitive strategy becomes necessary.
I will showcase this after having gone through two popular marketplace strategies.
1. Bundling and unbundling
Industries are in a constant flux, between unbundling, where niche product are created for consumers and bundling, where different offerings are aggregated. While a niche service/product offers a higher quality to the end consumer, aggregation brings efficiency in selection on one side and cost savings on the other, (as the firm offering multiple products will naturally benefit from economies of scope).
Much has been written about this topic, the Unbundling of Craiglist being the most cited example. The diagram below illustrates this extremely well.
Recently there has also been a great piece written on the current opportunity for the unbundling of Reddit by Greg Isenberg.
Currently a lot of media is being bundled, think of news via Google, Music on Spotify or films on Netflix: all these will present in time opportunities for unbundling. Some of those opportunities are already present. Think Mubi for high quality films, DIUO for live music or Pocket and Flipboard that deliver curated news. Even the mighty Google is being disrupted by, among others, vertical B2B search engines life ThomasNet and Drugdu.
Multi-category marketplaces are also called horizontal, (Craiglist), while single category marketplaces are called vertical – (these are the unbundled ones).
2. Managed vs unmanaged
Over the past 5-10 years we have seen a number of players that have come along creating generation 2 and even generation 3 marketplaces, which decreased margins by adding services on top like workflow integrations, payments, curation, distribution and occasionally even owning some supply.
These are the so-called managed marketplaces and they are a natural evolution of pure match-makers. In a competitive market high margins will always attract new entrants who are ready to lower margins by offering users a better value proposition. Think of Amazon: it has developed a logistics infrastructure specifically to create higher stickiness to its marketplace.
In certain sectors where margins might already be low, disrupting players have come in offering additional services to deliver a better user experience and to justify a higher rake (fee). So we have two potential strategies here:
A. Gaining market share by offering a better, (and higher-touch), user experience for a particular user segment at no extra cost
B. Gaining market share by offering a better user experience to a user segment that needs it – charging a premium
Different markets will dictate which approach is the winning one.
Personally, I am not too hung up on the definition of managed vs unmanaged, as one model can transitions into the other and vice versa: what matters is being aware of the underlying concept.
So what?
Naturally the two trends overlap quite well. Often the unbundling opportunity presents itself when a large aggregator, which will by default need to standardise the service, will fail to deliver a strong value proposition to different user segments within the platform. At this point new players come in offering a new tailored experience to a specific user segment and enriching the service of top e.g. improving trust, matchmaking, payments etc.
There is nothing new here.
The problem is that as these two strategies, that often work together, can be implemented across all sectors, how do we decide to opt for one vs the other? When are there unbundling opportunities? How does a company create a moat?
While unbundling can be a very successful strategy, success is by no mean guaranteed. Many are familiar with the unbundling of Craiglist diagram, fewer are familiar with the diagram below: one I find even more insightful (credit to Josh Breinlinguer).
The irony is that at the opposite end of the scale we have seen new players succeed in these markets without a marketplace model. It’s easy to imagine that once a marketplace has added trust mechanisms, payments solutions, logistics, curation etc, then by owning the supply they deliver an even better service.
This leads to two major implications for designing a new marketplace venture:
1. The requirement of understanding in depth how network effects work and how they add value, (e.g. recently I have been approached by a company that had set-up a marketplace for cleaners, while I see how it may work for some user cases, I also know that in most cases a standard agency will deliver a similar if not better service)
2. The requirements of understanding how a market works, which to an extent means defining the market, because joining supply and demand successfully does not equate to success anymore. If network effects have a limited impact, one needs to understand the key success factors.
Hard work and good execution are critical for the success of any start-up, but they are not sufficient. I have also seen a fair number of scale-ups with good traction fail either because their business model did not work at scale or because they clashed with some industry dynamic that made it impossible to grow further.
What has led to these failures? What leads to success?
A standard lens to look at this is that of frequency vs transaction value.
From a unit economics perspective, success in areas where frequency is low and purchase value is low is extremely difficult / impossible. That is why bundling many niche markets together is strategically the right thing to do.
The chart above, (courtesy of Sameer Singh), showcases how there are cases of success across all quadrants. That’s because the real equation to solve here isn’t frequency vs basket size, but CAC vs LTV (customer acquisition cost vs lifetime value).
This can be solved in a number of ways.
Decreasing CAC
This can be achieved by creating pull via virality or community on one or both sides of the marketplace – or through innovative content loops e.g. Etsy. Fundamentally any innovation that drives down customer acquisition dramatically on one side of the marketplace will unlock new opportunities.
Increasing LTV
The two main tactics here are:
1. Cross selling other services (e.g. bundling services that serve adjacent needs in consumers’ minds, like Thumbstack for home services and repairs)
2. Upselling one side of the market from software or hardware.
These tactics can also be used simultaneously.
Upselling from software has been very popular recently, where the software is the first sell and the marketplace is layered on top – this enables for multiple revenue streams, (SAAS fees and marketplace fees), and can create lock-in, improving retention and LTV. Multiple players have also used software as a product-led growth, giving it away for free and monetising later on the marketplace.
(Upselling from hardware is difficult, but can be extremely lucrative e.g. Apple and Tesla. Facebook is trying to this with Oculus)
As mentioned above: when network effects are weak, it’s important to understand what the key success factors in a market are – because the nature of network effects can’t be altered.
In some cases this will come down to business model innovation like:
- Monetising data
- Data capture to offer a better service & lock-in (increasing LTV)
- Implementing a subscription model to subsidise a more aggressive CAC
- Adding a SAAS revenue stream (see above)
In other cases success will come from better understanding the value chain and disrupting it – possibly bringing in new entities into the marketplace. This bring us to the third strategy for disrupting NFX.
3. Value Chain Disruption (or restructure)
Let’s take the home rental market as an example. A snapshot view will tell us immediately that the incentives in the market are not aligned. (The market of reference here is the UK)
Estate agents make money through new contracts, so they are incentivised in filling properties without caring who lives in them and if they get a good service – if a new tenant leaves they will monetise by finding new ones. Agents also make money by managing properties, but most landlord don’t buy that service because they don’t trust agents. This is ironic because agents are better placed to offer the service and most landlord would rather not have to cater to tenants.
These are 2 clear signs that the market is broken
When we look at the value chain of the market we see the following:
(Wardley maps, used below, are a great tool for this)
It’s clear that the agent has been very much commoditised (as it’s in the bottom right in the chart above).
This is mainly due to online aggregators which have made estate agents much less valuable. Estate agents are still used, but they are seen as a necessary evil.
So how does one disrupt the property rental market?
One could cut-out the middleman, avoid the “necessary evil” and improve market efficiency. Yet this strategy has been tried many times with very limited success. Today above 95% of transactions are facilitated through an estate agent. This is due to logistic and regulatory reasons which make it complicated to create a P2P market.
Another option could be to include and elevate the estate agent.
Imagine a market in which agents, enabled by an innovative solution, are capable of offering a better service where:
1. Software improves the billing, switching and legal document processes
2. A managed service improves the moving task for tenants
3. A managed service allows agents to earn more maintenance contracts
4. A marketplace solution offers added-value services on-top for tenants e.g. local services, discounts, local events and entertainment
5. The Bills management area plus rent will represent roughly 40% of household spend: this positions the firm well for playing in the FinTech space.
The impact of the above would be:
Tenants: would be happier for receiving a better service and could benefit from added services on top, (in a world where consumers are increasingly expecting everything on demand this has good potential). They would also be more confident with a reliable maintenance service offered. Ideally they would feel more at home in the property they are renting and in the neighbourhood they are living in.
Landlords: would be attracted to the service by standardised and improved logistics and not having to bother with tenants, bills, payment collection and home maintenance.
Agents: agents would have the opportunity to reposition themselves higher in the value chain, delivering more services, offering more value and being able to justify higher fees.
Marketplace: This solution would be part marketplace part platform (for the difference see my previous post). It can win through the combination of three factors:
1. The company would be in the position of operating at scale in offering certain services like plumbing and moving and general maintenance. Here owning the demand and leveraging economies of scale will allow for better and a more cost-efficient service than competitors.
2. A SAAS based solution can add significant value to estate agents, tenants and landlords alike, justifying a significant fee
3. A marketplace connected to the software will increase stickiness, create new value and generate a strong local network effect, which should translate into additional revenue and defensibility.
The solution would look something like this:
I think there are two takeaways from this example:
1. Designing for the ecosystem instead of trying to disrupt it at all costs can be a better strategy (in this case working with the agents instead of against them). This means designing for attraction.
2. Aggregation and price comparison are rarely the only dimensions in which a firm can add value. While everybody needs a house, what they ultimately want is to feel at home, (I highlighted this in the diagram above). Offering this kind of value-add creates the kind of intangible value that justifies higher fees
I particularly like this example because once we start analysing the market, it’s apparent that the dynamics within it are broken, and yet most firms are trying to disrupt it in the wrong way. While most large aggregators are competing with each other, this example showcases how a new market entrant can not only steal market share, but also be more defensible.
Watch this space and if you have any questions don’t hesitate to get in touch here.