Building hyper-local multi-country platforms

How liquid and how dense is your supply and demand? or: Not all markets are created equal.

What is a “hyper-local multi-country platform”?

Over the last couple of years I have had the privilege of working with several international marketplace companies that operate — what I call — a “hyper-local multi-country” model (1) and I was able to support them in expanding their business. Among those startups are: Eversports of Vienna, Paris-based StarOfService, Berlin-based Lieferando (now a part of the TakeAway group), Storefront from New York, Amsterdam-based Studocu, London-based Syft and, of course, Uniplaces headquartered in Lisbon.

More famous and well-known examples of companies operating under that model are Uber, Booking.com, GetYourGuide and AirBnB.

What all these companies have in common is that they need to build liquidity from scratch in every new market they expand into. Often, it’s not just limited to one side of the marketplace, they have to build both supply and demand, on a (hyper) local level, in order to launch a new (local) market. Why is that?

In all of these cases, at least one side of the marketplace, i.e., the supply or demand:

A) has a strong local component and is thus “illiquid”, i.e., it’s completely immobile (totally bound to the location) or, at most, only locally mobile/liquid (“mobility”);

B) is offline, i.e., the purpose of the marketplace is to make it available (or even better, bookable), online (“nature”).

Analysis of selected HLMC platforms (2)

Another reason is that network effects between the various hyper-local markets do not exist, are weak or, at best, take a very long time to build.

• Lieferando: In most cases, a consumer will use the platform to order food for delivery to his home or office. There might be a benefit of using Lieferando outside of your home city when traveling (e.g., business travel) but this only represents a small portion of the overall food delivery market.
• StarOfService: If a consumer is in need of, for example, a carpenter or tutor, he/she will probably choose someone that is located nearby since travel cost and time make it inefficient beyond a certain distance. In addition, a service professional has a strong incentive to limit his/her services to a certain area for the same reasons.
• Eversports: Sport fanatics will use the platform to book a venue that is close(st) to them (i.e., their workplace or home) but there is no benefit in having a lot of choice across the country if the platform does not offer what he/she is looking for in the same area. Similar to Lieferando, it will only be important if a sport enthusiast wants to continue to do sports while traveling, or in the case of someone moving.
• Uniplaces: A student will be looking for an accommodation in a particular city or even district, at best closest to the university where he/she has been accepted. There is no additional benefit for him/her if Uniplaces offers a lot of options across the entire country far away from the actual study location.

In these four examples, there is clearly more benefit in focusing on depth and breadth of supply on a local level (e.g., multiple sport categories and multiple venues for each category in one city), rather than a wide and deep offering in aggregate across multiple locations situated far from each other (e.g., in separate cities).

To sum it up, for HLMC marketplaces local supply- and demand-concentration, i.e., density, is key. That is why, depending on the structure of the respective market, many of these companies choose a city-by-city expansion strategy when growing their business internationally.

In contrast, platforms like Upwork, Fiverr or 99design have very “liquid” supply and demand. The services offered on these platforms can be rendered regardless of location, i.e., remotely, and thus the entire process is 100% online. In fact, language is the only potential barrier, but in today’s globalized business world where everyone speaks English, this is manageable and does not represent an obstacle. As a result, a freelancer in India listed on Upwork can offer and deliver services to a European-based company without the need to travel.

Now that we understand the definition and general characteristics of an HLMC marketplace, let’s deep-dive into particular challenges when building such a business.

Organizational structure: local vs. central

One of the major challenges when expanding a hyper-local multi-country (or better: multi-city) platform is finding the right answer to the question: Which part of the operations should be done centrally, i.e., from the headquarters, and which part should be done locally “on the ground”, i.e., in the respective markets, by a dedicated local team?

While it always depends on the individual business and the structure of the market it operates in, I have developed the following rule of thumb: be where your clients are (3); if most relationships that are managed through the marketplace have a local component, most of your operations should probably be local, too. For example, in the case of Uniplaces, the following (contractual) relationships are managed through the platform:

• Student (Tenant) <> Landlord: subject to local law (e.g., for an international student moving to, and studying in Germany, the applicable law would be German, as the landlord AND the tenancy agreement are in Germany)
• Student (Tenant) <> Uniplaces: independent of local law, depending on where Uniplaces is incorporated (i.e., the UK).
• Landlord <> Uniplaces: subject to local law (e.g., for a German landlord with a property in Berlin German law would apply)

In the above example, two out of three (contractual) relationships that are facilitated through the platform are subject to local legislation, and thus have a strong local component, hence it makes sense to also focus your operations (such as sales, customer success, account management etc.) locally.

Product

Whereas it can make sense to build your operations locally, it often does not make sense to “decentralize” your product department and build x types of products, i.e., a dedicated stand-alone product for every individual market. This would simply consume too many resources, especially for a startup, and would make the entire development process very inefficient. That said, an HLMC platform often requires certain product features or adjustments that are specific to one particular market, while the company is attacking multiple markets at the same time. This creates a dilemma for the product department: Where to focus the development resources, and which features should be developed? For example, customers in one market might request a feature that no-one else (in other markets) will find useful.

In my view and experience, it’s important to leverage the “local knowledge” in the organization: Provide tools and decision power to local managers in order to influence the product roadmap in a meaningful way. At the same time, make sure that your product team understands the local needs and always has “an ear to the ground”. (4) It´s crucial that product and local teams exchange on a regular basis, understanding each other’s challenges and listening to each other.

Sales approach and organization

Not all markets are created equal; make sure you understand the market structure of each individual market that you plan to enter. You can’t always apply the same approach and strategy that you’ve used previously. For example, the supply structure in one region might be characterized by a highly fragmented, extreme long-tail of small accounts, whereas in another region you have to deal with a much more concentrated structure, in which only a few large accounts control a significant part of the supply.

The first market would require a “no-to-low-touch” sales approach (e.g., centralized telesales), whereas for the latter it might make sense to choose a “high touch” (decentralized) sales approach, where individual key account managers are responsible for field sales. Hence, you can’t simply “copy and paste” your sales model, you need to understand every market and its structure in detail, in order to choose the right strategy market-by-market.

Once you understand the market structure, you need to decide how to build the sales department. On a local, country level, do you choose a centralized or a decentralized, hyper-local sales approach? For example, in the case of Uniplaces, a foreign startup expanding to several cities in the German market at the same time, is it better to build a centralized German sales team that is based in one German city from which it targets all other cities at the same time, or build a decentralized sales team with local sales reps based in every target city?

Based on my own experience, it makes most sense for the sales organization to follow the organizational structure of your target group. If most of your (enterprise) clients have dedicated account managers for every region it makes sense to build the structure of your sales organisation around the same model. Or, simply put: be where your clients are!

If your target group requires a light touch approach, before building your own internal team you can also “test the waters” and outsource initially, or even better: “near-shore”, to an external call center within the particular market into which you plan to expand, and then insource once you see that it’s working. This does not require such a big upfront investment, and usually allows you to test the market in a faster and more efficient way.

Internal Communication

If you’re operating an HLMC platform one of the biggest challenges is often managing and fostering internal communication between the various local teams as well as the headquarters. You often end up with distributed (small) teams in various regions and a (larger) headquarters with central functions. It becomes even more complicated once you are active in multiple time zones. How can you make sure that information flows smoothly between all of the teams and that all employees feel a part of the same team and company?

There is no easy answer to that question but there are a few best practices that you can follow:

• Embrace asynchronous communication: Use tools (beyond email) that help you to communicate asynchronously with the rest of the company, e.g., Slack, Wikis and/or Basecamp.
• Institutionalize knowledge exchange: Set up regular, recurring appointments with a clear format and agenda in order to facilitate ongoing knowhow exchange between markets, and to motivate people, especially to share mistakes and failures. Ask questions such as: What was the biggest mistake that I made last week/month? What was the biggest win or lesson learned?
• Set up a kick-ass infrastructure: Use top notch video and audio conferencing solutions and don’t try to save money on the technical equipment! If you put your (local) teams in professional coworking environments like WeWork, you usually don’t have to worry about these things.
• Make people travel: Bring people who have the same functions together on a regular basis — both offline and online! Make new joiners spend enough time in the headquarters at the onset.
• Have fun together: Hold a big company party, bringing everyone together once a year, in one location (where nobody can escape)! ;-)

M&A - buy vs. build

When expanding into new, international markets you may come across existing local players that are already active in the target region. Some of those that have a relevant size (5) might be interested in a) getting acquired and thus in b) entertaining M&A-talks. Those discussions are often very time-consuming and the outcome is difficult to predict, making it hard to judge whether it’s worth spending time on it. Furthermore, you have to ask yourself the (in)famous “buy-or-build-question”: How much time and effort — translated into financial investment — is required to get around them and win the market vs. how much would you need to pay in order to acquire the other company? How much could you accelerate by buying a competitor vs. developing the market yourself? This question alone deserves a blogpost. (6)

At this point, I would like to highlight a few examples of HLMC marketplaces that have consciously chosen an aggressive M&A strategy in order to expand internationally. Sometimes the circumstances force you to do so because local competitors have already reached a critical size that makes it too costly or too time-consuming to get around them. That’s one of the reasons why successful marketplace businesses such as Allegro Group, Delivery Hero, the TakeAway group, and or Helpling choose to acquire one or more sizeable competitors. Delivery Hero probably serves as a European posterchild example for this strategy: According to crunchbase, the company acquired 14 businesses and made 8 investments (into competing/adjacent players) between 2012 and 2017. Such a strategy obviously requires significant funds and an experienced, in-house, M&A-team in order to prove successful.

Hiring

When it comes to hiring, the particular challenge of HLMC platforms is the remote nature of the various teams; for example, when you hire new sales reps for a particular market, there might be no obvious reason for them to spend time in the headquarters. I strongly believe that it’s very important for them to spend significant time at the beginning of their new job with the central teams, in order to soak up the company culture and subsequently carry it to the local hubs.

Besides, your internal recruiters might be organized centrally, so hiring lies primarily with the local teams. It should be conventional wisdom by now, but if you want to do yourself a big favor, use a cloud-based Applicant-Tracking-System (ATS) like Lever, Greenhouse, Workable or Recruitee. It allows everyone to stay asynchronously connected, including central HR and local teams, and facilitates a fast and smooth hiring process across the distance.

The “local component”

As defined and explained above, the strong, offline, local component of HLMC platforms creates two main challenges:

1. Those businesses tend to be more “operations-heavy” and thus more dependent on execution excellence than other tech-startups.
2. At the same time, the local factor makes them prone to the so called “bypassing issue”: Since both sides of the platform — supply and demand — are typically in the same location at one point, they could theoretically transact directly between each other and hence bypass the platform.

A possible solution to both of these challenges is to really focus on product development, and the added value that the platform creates for all involved stakeholders. In addition, try to automate as much as possible. Follow this highly simplified rule of thumb: better to hire one additional developer to automate tasks than three more people in operations as a “quick fix” that was initially only meant to be a temporary solution but, in the end, just creates more “operational legacy” and stays forever.

Timing

When is the right moment to start expanding your HLMC? Generally speaking, you should figure out the right model and product-market-fit (PMF) in your initial home market — be it on a city, regional, or country level — before expanding aggressively into other markets. That said, and as previously mentioned, every market has its own challenges, hence they might require certain adjustments, so you need to find local PMF for every new market you enter. Ideally, you can show a working model (in terms of demand-/supply-acquisition, revenue/monetization, unit economics, growth patterns etc.) in one particular category, for one specific target group, in one local market, before horizontally expanding into other categories and new geographies. For example, in the case of the Dutch operation Treatwell, another HLMC platform, they would first need to define and perfect serving hairstylists in Amsterdam before trying to replicate the approach in other cities and other categories, such as massage, for example.

Why build a hyper-local multi-country platform?

Considering all these challenges and obstacles when building HLMC platforms, you might ask yourself: why am I still excited about this type of business at all? Or, as an entrepreneur, why should I even bother building one?

The answer is relatively simple: Overcoming all these challenges while constructing the network creates and unlocks tremendous future value. The value itself lies in the network that you are building over time. You might not be able to harvest the fruits from it right at the beginning, but owning those local markets is an asset itself, that creates a defensible moat around your business which is hard to replicate for new market entrants or other competitors. Thus, operating under such a model allows you to increase your long-term defensibility and creates a sustainable competitive advantage. At best, you are the standard “operating system” of the local market. Eventually, it also creates greater monetization potential once you actually own a market and close it against competition. Especially in combination with an SaaS-enabled marketplace-model, this can prove very powerful, and allow for greater revenue potential.

Long-term you may also be able to create cross-market network effects like Uber or Airbnb. Whenever I travel, I can use these apps almost anywhere in the world. Of course, this takes time and money, but it is not only a very compelling business-case but also a promising investment-case.

I am convinced that Europe is a great breeding ground for HLMC platforms, because companies have to expand quickly internationally, in order to get to a meaningful size and be able to build a “venture case”. On top of that, the European Union facilitates cross-border business, harmonizes standards and makes it easier to expand. That is why European founders are typically not afraid of internationalization and often expand their business early in the lifecycle of the company. (7)

If you’re working on an HLMC platform or thinking about building one, get in touch with me! Any feedback about your own experience in operating under such a model will be highly appreciated in the comment section below.

Many thanks to everyone who helped me writing this blogpost, in particular Ruben Timmermann of Springest and my buddy Phil Noack for reviewing an earlier version of it.

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

Footnotes

(1) In this article the abbreviation “HLMC” is used for “hyper-local multi-country”.

(2) Included as counterexample only for explanatory reasons (so Netflix or Upwork are obviously no HLMC platforms!)

(3) In this context “client” means that side of the marketplace that has the money and pays the marketplace

(4) Especially important in a “low touch” sales environment — this is your only way to get local feedback!

(5) “Relevant size”: They have crossed the inflection point and have reached critical mass e.g. their reach to the relevant target group exceeds 50% market share.

(6) Take me up on that! ;-)

(7) Perhaps it’s also the parts of our DNA that the vikings passed onto us.