Cisco’s launch of Connected Mobile Experiences (CMX) Beacon Point today is probably the most significant event in the Bluetooth beacon ecosystem since Google announced Eddystone.
The Beacon Point virtual beacon technology is amazing. There are significant business benefits. As important, the biggest networking company in the world has chosen to compete in the Beacosystem. This has implications for the other 500 hardware, software and services companies already operating in the market. This latest milestone is worthy of serious analysis as there are opportunities and threats for Cisco and the rest of us.
Beacon Point enables the location of smartphones to an accuracy of 5 feet without the need to place hardware within reach of visitors to a venue. It’s being targeted initially at wayfinding and proximity alerts in retail and hospital applications.
The system allows up to eight virtual iBeacons to be deployed and managed via a single Beacon Point antenna, using Power over Ethernet (PoE), so no electricians required for deployment.
With Beacon Points secured at ceiling level, staff no longer need to be dispatched to hide beacons under shelves. Virtual beacons, or vBeacons are positioned and managed via the cloud using a web interface that allows a beacon to be placed on a floor plan with the click of a mouse.
No worries about running out of battery. No risk of beacons being stolen. No need to deploy staff to do firmware updates or to move beacons when the environment changes, and no worry about what will happen if your beacon vendor goes out of business or gets acquired.
Cisco is likely to be the one doing the acquiring.
What Is It?
The Beacon Point looks like a large Wi-Fi access point that could be the sibling of Cisco’s Hyperlocate Wi-Fi location or “Halo” unit. It is actually based on a different platform that uses an antenna array with 8 elements. These accurately “form” beams of Bluetooth signals that radiate out from the array to handsets in its vicinity. Each of the 8 elements broadcasts a different UMM (UUID, Major, Minor number). A combination of AltBeacon and iBeacon packets leverage iOS’ ability to “wake apps up from the dead” in a battery efficient manner when a proximity event occurs.
For Beacon Point to implement its virtual beacon functionality, developers need to use a Cisco proprietary API. This takes over when the first Beacon Point packet is received. The SDK uses a cloud service which Cisco provides to do the location computations.
The service looks for multiple Beacon Point signals and does some trigonometry and machine learning to figure out the phone’s location in the X and Y axis. You need 3 Beacon Point arrays to be visible to achieve 5 feet of accuracy. The cloud service then maps the estimated X/Y coordinate to the vBeacon zones setup via the web console. If the phone is within the bounds of one of those zones, then the vBeacon UMM is used to pass control back to the app.
From iBeacon to vBeacon
A single Beacon Point can cover an area of 2,500 square feet which allows for 8 vBeacon zones. Given an average Wal-Mart is approximately 150,000 square feet, this would require 60 Beacon Point units enabling up to 480 vBeacons. Beacon Point doesn’t enable iBeacon’s most accurate region monitoring, “Immediate”, which generates a trigger when the phone is a few inches away from the beacon, ideal for enabling a trigger when the phone is next to a specific product, but it does support “Near” regions of 5 feet radius, suitable for triggering proximity alerts relevant to categories of products in the store, e.g. cat food or light bulbs.
vBeacons can be assigned to a space with a point and a click anywhere on a floor plan. This offers a level of flexible control that can be challenging to implement with physical beacons that need to be placed on supporting structures. There may not always be a pillar or shelf in the right place to fix a physical beacon and implement the use case. vBeacons address the vulnerability that physical beacons have to being tampered with.
They also have less negative impacts on the look and feel of a display (given that they are invisible). Where the real benefits come into play is when beacons have to be moved or adjusted. This might occur at an exhibition center every few days when a new event begins, or when a retail store layout is “reset”, for example at the beginning of season or for occasions like Halloween or Christmas. The operational overhead and execution challenges of coordinating the relocation of beacons across 5,000 stores, either with a fleet of service technicians or instructions to staff on site, is a massive cost and is liable to incur errors. With Beacon Point, all of those changes when the aisles and product planograms are reset can be executed from HQ, where those layout changes are often prescribed, at least in the case of larger more tightly managed retail chains.
All of this assumes a level of control and order that may be true for the most organized and disciplined retailers. Categories and the layout of products are managed scientifically from HQ, where the planogram that specifies the layout of products on shelves closely matches the “Realogram” which records what’s really being done. Some retailers may delegate that authority to local store managers, which doesn’t exclude the use of Beacon Point but weakens some of the arguments in its favor.
Cisco is known for charging premium prices, which has been problematic for it in verticals such as retail. The cost of Beacon Point may reinforce that perception. The hardware and services are priced at the high end of what we have seen in the Beacosystem. Beacon Point units list for $695. The corresponding 8 physical Kontakt.io beacons would cost $160. If you added Kontakt.io’s Bluetooth to Wi-Fi Gateway to enable remote management, that would be another $89, for a total of $249, less than half the cost of Cisco’s bill of materials. Kontakt.io’s double battery product with a claimed 5 year battery life would bring the beacon cost of 8 beacons up to $216. Kontakt.io’s Gateway is USB powered so would require some cabling similar to the Beacon Point.
Cisco also charges a $150 annual cloud fee per Beacon Point, which supports 8 vBeacons (for a 2 year subscription). Kontakt.io doesn’t charge for their cloud product, but vendors such as Pointr, Nearby and Indoo.rs that provide location and wayfinding in the same category as the Beacon Point cloud service, do charge for their location services.
If you were to max out our Beacon Point Wal-Mart configuration, that would be a list price of $41,700 with cloud services that cost $9,000 a year for a single store, over $200m in capex for every store (5,000+ stores) and $50m a year in cloud service and software support fees (after the first year which can be free). Of course Wal-Mart may not need a maxed out configuration and would never pay anything close to list price, but there will be a limit to what Cisco can do to reduce the hardware price and that means that there better be some significant revenue lift in the Return on Investment calculation to justify a national deployment.
CTOs Choosing Cisco versus Hamilton!
CTOs have been happy to give a “shot” to “young, scrappy and hungry” startups to do their proof of concepts (to borrow a phrase describing Alexander Hamilton, taken from the musical). CTOs benefit from the startup’s agility and expertise when the risks don’t matter so much, but when it comes to doing large roll-outs, they may look fondly to the high tech “monarchy”, even if they have to pay more in “taxes”.
For a CTO who wants to centralize control, who thinks about total cost of ownership including re-deployment costs of beacons and already has relationships and SLAs with Cisco, they may relish the opportunity to consolidate the suppliers they use. Their life is complicated enough and being able to use the same vendor across Wi-Fi and Bluetooth networks may be a great way of bringing this “rebel” technology into their “empire”. That’s not to say every CTO is King George bent on reigning in control, but having the same tools and staff able to manage all their wireless infrastructure is going to make a lot of sense to companies already using Cisco tools in their Network Operating Center.
Beacon Point may be “revolutionary,” but it isn’t Cisco’s first venture into the beacon ecosystem. Its sister Wi-Fi Hyperlocation product, the RM 3010, has a Bluetooth radio that can be used to manage real physical beacons versus virtual ones.
Beacon Point isn’t the first virtual beacon product either. Last year SmartFocus announced a virtual beacon product that has since disappeared from the market. Clearly Cisco have a level of credibility that will allow them to be “in the room where it happens,” that is, when products are evaluated by big customers in a way that the UK startup couldn’t hope to rival. Accuracy is key to the virtual beacon concept working. If SmartFocus was working with commodity hardware, it would have been difficult to achieve the level of precision necessary to make their technology “revolutionary.”
“Wait For It” – It’s far from over for Beacosystem Startups
There’s still plenty of opportunity for the other players in the Beacosystem. Cisco may be the biggest network infrastructure player, but Zebra and Ruckus are arguably stronger in the retail market, so the benefits of consolidating Wi-Fi and beacon infrastructure with Beacon Point don’t apply if your infrastructure comes from those vendors.
Virtual beacons have major advantages over physical beacons, but they do require more work when they are first deployed. Beyond the laying of cables to power Beacon Point, a floor plan is required and the location location of each unit needs to be precisely recorded on that plan. This is an unnecessary overhead if all you want to do is record when customers have arrived in the store. That can be achieved much more simply with a single beacon, stuck on a wall at the store entrance.
For many applications, the high level of precision that Beacon Point offers is not required. While the Cisco sales force doubtless are under pressure to deliver more accuracy than a generic Wi-Fi only solution can deliver, the vast majority of use-cases being deployed today don’t require precision. Companies like inMarket, who are one of the few beacon solution providers generating cash from their business, don’t require high levels of precision, generally they just need to know if you are in the store or not. That’s not to say that precision doesn’t have value. There is a certain amount of “chicken or egg” about this. Until there is a credible source of high precision location infrastructure, there won’t be many apps that use it.
For customers committed to integrating with Google’s Eddystone UID and EID protocols, Beacon Point may not be the right choice currently. Those frame types are not supported yet, although that would be an obvious next step on Cisco’s roadmap.
Since Beacon Point is dependent upon use of its SDK by the apps that need to see vBeacons, it’s hard to see how Eddystone URL can be supported by Beacon Point unless Google adopts Cisco’s SDK. Cisco and Google have an Aaron Burr and Alexander Hamilton style partnership: when faced with a common enemy (e.g. England/Microsoft), they may collaborate, but it’s hard to see how this alliance would extend to the Beacosystem.
Beacosystem startups still have the advantage of agility, focus and expertise in the field. The Beacosystem is still moving fast and it remains to be seen how aligned and committed all Cisco’s functions are behind this new product.
Last but not least, asset tracking is one of the most lucrative classes of use case for beacon hardware vendors. Attaching beacons to people, parcels and pieces of equipment, is driving more beacon sales than tracking the movement of phones. Beacon Point doesn’t seek to address those use cases. The way Cisco’s product works today, the smartphone is an essential element of the architecture. It is the smartphone that is monitoring the beams of Bluetooth signal from the Beacon Point arrays.
What does Cisco have to do for Beacon Point to “Stay Alive”?
Like an operating system platform, beacon platforms’ survival depends on apps. Since Cisco’s APIs are proprietary, they can’t rely on the existing apps ported to generic iBeacon and Eddystone APIs, they will need to recruit developers. Having big Cisco sized deployments will naturally pull developers in their direction, but that’s probably not going to be enough.
The core of the Beacon Point product appears to have been sourced from a Cisco partner called Mist. Those familiar with Cisco, may look at Mist’s executive team and think “I Know Him”. Their COO and co-founder, Bob Friday, used to be Cisco’s CTO. This is positive because it means there is a nimble and expert source for the technology.
“What Comes Next”?
One signal that will make Cisco’s commitment abundantly clear is if and when they integrate Beacon Point and Hyperlocate into a single hardware platform. Being able to unify Wi-Fi and Bluetooth analytics would be a formidable offering. In fact it would “Blow Us All Away.”
While this analysis may have too many show tunes references to be considered completely serious, it’s clear that what Cisco has done deserves respect. They have tackled the total cost of ownership that impacts large beacon re-deployments. They have gone from providing infrastructure to passively monitor other vendor’s beacons, to creating a new beacon category, the virtual beacon.
There are enough gaps and opportunities left that most incumbents should see this development positively.
While some beacon companies may find that Cisco is now a formidable competitor, the entry of this giant into the market will do more to grow the pie than the downside of having to share it. For conservative customers who have been holding back on their investments to see if beacons are a fad, this vote from #54 in the Fortune 500 is a powerful signal that the Beacosystem is here to stay
Watch our Cisco YouTube Interview Discussing Beacon Point