Downstream Space Wars – When everyone is Super

A war is looming in the downstream space sector.

The adversaries are the Earth observation platform providers.  The battle ground is “content”. 

The notion of “content” shaping the market success of new technologies is not new. In the 1970s the range of movies available on VHS helped it beat Betamax, and when several film studios committed to Blu-ray in 2008, it was the death nell for HD DVD.  Today, similar content battles are being fought between Netflix and Disney, or Xbox and Play Station.  “Our future largely lies in exclusive original content,” said Netflix in its latest earnings report.

The same is true for the geospatial platform providers.  The “content” is geospatial data, fuelled by the explosion in new commercial Earth observation satellites. This is also the inspiration for predictions that the global remote sensing services market will double over the next 5 years from $10.68 billion to $21.62 billion.  Such growth is driving the need for exclusive content so that the platform providers can compete in what is fast becoming a very crowded market.

Let me recap some history, and terminology, for the newbies.

Some Background

Traditionally, geospatial data was collected by people — lets call them surveyors — who made a map product, sometimes from ground measurement and sometimes from aircraft. Some institutions or companies would do all of this in-house.  Most of them simply bought the map products from specialist survey companies. Virtually every modern business can benefit from better decisions based on improved maps.

Then along came desktop geographical information systems (GIS) so that various map layers could be combined together — decision making using geospatial data got a whole lot smarter, and a whole lot more complex.  In many businesses, GIS is at the core of their decision making, which they did with teams of GIS experts and paid large fees for software and hardware to crunch the data.  They also still relied on surveyors to collect the data and provide the necessary map layers.

Two things have shifted the goalposts in the last 10 years.  First, the proliferation of broadband internet.  This allowed for rapid transmission of very large geospatial data files, which moved many GIS companies like ESRI to shift to offering their software “as a service” (the SAAS model). SAAS meant the software and the number crunching happens off-site, on someone else’s computers (on The Cloud somewhere), rather than with the customers. The customer gets their map information via a web browser, reducing the need for (expensive) technical staff, (expensive) hardware and (expensive) desktop licences. 

The second shift is the enormous proliferation of Earth observation (EO) satellites. 2018 is barely half way though and already 54 EO satellites have been launched, 30 of them commercial.  Compare that with 2008, when only 20 EO satellites were launched, and only 5 were commercial.    There are now over 500 EO satellites in orbit collecting some kind of data about the Earth, and that is only going to grow, with companies like Iceyeand OMSpushing for large constellations of many tens of identical satellites. 

For customers who already have a budget line for geospatial activity this means they no longer need to rely solely on the services of survey companies, since data is now ubiquitous. It doesn’t matter where you are on this world, there are satellite images available within a few days, sometimes for free. Within the next 5 years, there will be so many satellites[1]that everywhere on planet Earth will be imaged every few hours.

Now, I realise I haven’t got to platforms yet. Bear with me.  Lets first look at portals.

Portals: Gateways to Data

With so much satellite data flying around, its hard to choose which kind to use and then how to find it.  So, close on the heels of the data explosion has been the increase in online portals.

portal

I use the term “Portal” to refer to a web-based map interfaces that offer easy access to search, find and download the geospatial data of choice.  The portal service is usually free to search for data, but may require payment (one off, or subscription) to actually download the files.  The satellite launch companies usually have portals — Earth-i, Planet, Imagesat, Digitalglobe, all provide portals as a way to sell data from their own suite of satellites, or partner suppliers.

Many government agencies also offer portals — USGS for Landsat, Scihub for Sentinel — and some other organisations even bring together different types of data (e.g. Alaska Satellite Facility) but these are usually only other government satellite data sources, rather than commercial.

Portals: Analytics and Solutions

Now, portals still require a level of interpretation or analysis before the decision maker has enough information. It may provide some filters, or derivative layers, or simple analytics, but a portal still needs an expert user to make sense of it all.  So lets go up a gear, and bring in a wider variety of data. Imagine we can start to link the satellite data to stock inventories, supply chains, weather forecasts, environmental footprinting, commodity prices, localised political unrest, or any other level of data, the range of which is limited only by your imagination.  Now build that system so the output is less about imagery and maps, and more about situational awareness, audit reports, compliance checks, revenue forecasts, supply chain disruption alerts, or whatever shape or form the answer takes.   This is the ambition of “platforms”.  Platforms offer answers, not data.  Solutions, not imagery.  Indeed, a platform user might never see an image, or a map, and in fact may never wish to.

While the portals have been a necessary route to amplifying sales from commercial satellite companies, or amplifying use cases for government agencies, they are not as exciting as the concept of a platform. There will always be a sideline for portals, but serious users will use APIs to automate data discovery and download, not by hand via a website.  Platforms, on the other hand, are seen as direct routes to serving customer need, and have therefore also caught the imagination of investors for two important reasons.   Analytical platforms are not new, they are already core to many business intelligence systems, but the inclusion of huge amounts of new satellite data has introduced a new dimension to the potential of what platforms can provide.

First, and most importantly, the inherent scalability — re-selling analytical services multiple times on the back of the same data sounds an attractive business model (small margins, but large volumes by scaling globally, which is not difficult with satellite data, since if you can do it in location X, you can do it everywherewith only minor modification). 

Second, there is a relatively low barrier to entry in terms of technology — anyone with a server and a little know-how can set up a platform[2].  This means the start up costs can be relatively low for a big impact. Compare these platform companies and their investments:  GSi ($1.8m), Ecometrica ($2m), Bird-i ($3.6m), Rezatec ($4.2m), Planetwatchers ($1.8m); with the companies that are also building the satellites to collect the data:  Iceye ($54m), Planet ($183m).  Satellites aren’t cheap. Servers (relatively) are. 

Backed by investors with some very deep pockets, there are also some platforms, such as Orbital Insight ($79m) and Decartes Labs ($30m) who no doubt have every intention of scooping up all the market share that is available.

So what is the likely direction of travel in this sector?

Direction of Travel

My prediction is that the “Big Verticals[3]” will eventually dominate the downstream market. By a combination of inward investment and acquisitions, they will get their teeth into every stage of the process from satellite launch to client servicing.  For example, take a look at Maxar, a self-described “vertical space company”, which was originally MDA(and before that Macdonald, Dettwiller and Associatesfrom its origins in 1969). With a history of Canadian Government investment, and a track record that includes Radarsat (a pioneering success in commercial satellites), they were already in a strong position.  Then they acquired Orbital Optics in 2007, to build optical cameras,  and SSL in 2012 to build satellites, and then DigitalGlobe in 2017 to complete the portfolio of optical and radar high resolution satellites.  The DigitalGlobe acquisition came with their image portal, and on the back of that they also formed an in-company unit called Radiant Solutions who’s offering is a platform that offers tailored SAAS solutions.  

In Europe, the example would be Airbus, which ploughing the same furrow as Maxar[4]Airbus Defence and Space (formed in 2014 as a restructuring including EADS and Astrium) now includes in their portfolio Infoterra (geospatial analysis), SSTL (the original small satellite company in Surrey) and Ariane (launch vehicles).  As evidence of their intent, note that Airbus recently out-tendered EDINA for the coveted Jisc contract to supply map products to UK Academia.  That would seem small fry were it not tied to a larger ambition of providing platform-based SAAS services alongside. As the press release says: “Now, we’ve partnered with multinational earth observation specialist Airbus to develop a new, commercial quality geospatial data service for our members at highly cost-effective prices.”

Even these well-funded platform-based startups will one day have to be able to provide exclusive (or at least guaranteed) content, and so even they will have to start growing vertically.

The Small Verticals

While the Big Verticals are setting the pace, there are some outside bets — what might be called the Small Verticals.  Planet may catch up, especially if they merged with, or acquire, a company like Iceye to provide them with the radar capability they currently lack.  Earth-i is also an underdog, but all their graphs currently point upwards.  Blacksky, like Planet, don’t have radar in their portfolio[5]

In the business of meteorology you can see similar trends in companies trying to cover the entire supply chain.  Spire (“space to cloud analysis”), for example, or Orbital Micro Systems, with their ambition of 30+ cubesats offering unrivalled atmospheric measurements.

If the Big and Small Verticals dominate, what will become of the smaller platform providers? One suspects that those with something unique and scalable will be acquired by the Big Verticals to secure overall market dominance, or by the Small Verticals to ensure they lead in certain niche market segments. There will probably also be room for smaller artisan platform providers that service specialised markets that don’t need particularly exclusive satellite data, but do need tailored and bespoke platforms that onlyanswer their particular customers’ questions, but do so exceedingly well.  Given such an inherent lack of scalability, these will have to be high value clients with particular needs, such as for corridor mapping for network infrastructure (NM Group) or production forestry (Treemetrics), so that a high-margin-low-volume business model can succeed.   The key here will be understanding customer need and realising that they couldn’t case less how good your satellite data is, as long as they get quality answers to their questions. But at some point, these niche markets, which are currently largely subsidised by investment or government grants, may soon find that their market can’t accommodate the high-margin-low-volume model, no matter how clever the technology. 

When Everyone is Spatial, no-one is Special

While these artisan companies will still be dependent upon specialised content, and will differentiate themselves in the market based on their customer focus and specialised analytics, the big platform providers and the Big Verticals will be fighting it out over content. At the moment, all the players in the downstream are still focused on expanding the customer base for space-enabled platforms, and that is still a challenge for all of them.  But eventually there will be a point of saturation when the main platforms have all the space data they might need: full coverage, near-real-time (<3hrs) optical and radar data from satellite. My back of envelope calculation puts as early as 2023.  At that point, all the platforms will have essentially the same content, so that finding exclusive new content will be the battlefront of all the platform companies, because,  “When everyone is super. No one will be.” 

super
When everyone is super. No one will be.”

 


[1] Some that can even see through clouds, and in the dark, so can guarantee an image with every pass.

[2] Indeed, everyone and their granny are currently setting up platforms on a regular basis — I hear about a new one almost every month.

[3] The term “vertical” refers to the supply chain. 

[4] There are other Big Space Verticals, mostly in the communications sector. There are also some very big geospatial companies looking to make solution-focused platforms.  Hexagon Geospatial are driven by airborne and ground based content, but now have Airbus as a content partner, presumably on the space side.  ESRI is the stalwart of SAAS based GIS.  And dont forget the players in the East, such as Cyient, formed in India in 1991 and now a world-wide with 14,000 employees.  They are very engineering-focused and again reliant on ground and airborne content.

[5] Without radar, coverage is limited to cloud-free areas only. While constellations optimise the chance of cloud free imagery, it still ultimately limits the ability to guarantee image coverage at any particular time or place.

Advertisements

1 Comment

  1. Reblogged this on EarthEnable and commented:
    Very good blog post on ForestPlanet about the future of EO data, portals, and platforms. I also foresee near-real-time satellite imagery becoming available to the user community within the next 5-6 years. The boom of small-satellite constellations in both the optical (e.g. Planet) and SAR (e.g. Capella Space, IceEye) is already a game changer in the industry, and will cause more disruption in the EO data + analytics domains in the future.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s