When it comes to smart buildings- UI is not UX.

This article first appeared on Linkedin Pulse

If you think about the user journey of a commercial building- from commuting, arriving in the lobby, signing in, navigating, accessing the required floor, finding your host or colleagues, finding the right space for the right activity, putting your stuff in a locker and maybe ordering a coffee…. there is a whole bunch of user interfaces along the way. And this is the problem- UI overload.

I wrote earlier about how the native App is pretty much dead as this problem extends into all areas of life- too many apps creating too many user interfaces or even worse- notifications annoying us throughout the day. And in a building, not only does every system have it’s own app- there are also light switches, swipe readers, kiosks, touch panels and biometrics as additional user interfaces.

The missing part of all of this is User Experience design. A big part of the problem is all these various areas are isolated from each other- not just with technology but with process and ownership. Therefore, these problems can be solved by making small tweaks to the project management and roles and responsibilities. Let’s look at a few examples.

Tenants want more than what the building managers are offering

I recently had a meeting with one of the biggest tech companies in the world. I mention that not to boast but to point out how backwards building technology must be to a Silicon Valley company. They were planning for their new office and were frustrated with not having access to all the centrally managed elements of the building to include in their workplace solution. Elevators, secure access and other building services were simply not available to them. And the building managers didn’t seem to understand what the problem was. The elevator will have its own interface, you can use the bluetooth app for virtual building access cards and why do you even need to monitor the water quality?  

It got to the point where this tenant started building their own backdoors and workarounds. You say no to a company like this and they’ll simply use their tech resources to find a way. And their employees have allocated time to work on side projects- so this was a nice little challenge for the dev team.

What these developers discovered was something we also stumbled across at ACA: the building management systems and central building software is redundant. It’s basically unnecessary middleware that hasn’t changed for decades. Why not just bypass it and talk directly to the end-points?  And this is where the building managers and leasing companies get in the way. They just want to do what’s easy- use the systems they know- follow the same formula. It’s not an easy area to disrupt but conversely this makes it a soft target for disruption.

If you are interested in more of my thoughts on bypassing unnecessary middleware- the last part of my webinar (this link) covers that.

UX is not about technology- so don’t let your IT department lead the process

The  above example shows us that there is at least two big silos: the building and the tenancy. And this is why we might see multiple interfaces between the two. In the same vein, it’s also why there is double handling in other ways. For example- why do I have to sign in at the building reception- then do it again at the workplace. If we include security- sometimes we have  3 interactions before the host knows we are in the building. “Sorry I’m 10mins late- but I arrived 10 mins early!” There are many more silos we are challenged with across a smart building.

If we look at a single tenancy- a workplace- there are some consistent problems we see- and sometimes solve. I’ll lead with the solution: IT should not be making the technology purchasing decisions. A business unit should run this- with technical input from IT.

If you run everything through an IT decision process it will look something like this:

  • We need an x solution. EG we need a room booking solution

  • We’ll write the must haves and nice to haves requirements list.

  • We’ll get 4 products into a POC environment

  • They all have to pass our network and security audit first

  • Then we’ll tick off our requirements and the one with the highest score gets spec’d and we’ll put out an RFP to the vendors channel partners.

The problems this creates:

  • You are leading with a product: not thinking about the broader solution. If you scope the UX you might find you can solve this problem by removing technology- not adding tech. It's hard for IT to think like this. Also, you might be replacing something you already have with something that does the same thing. Sometimes the problem is not the tech- it’s the UX. And if you replace the tech to get the same UX- you are left with the same problem. This is fresh in mind for me as a common thing I hear is ‘we are not happy with our Condeco room booking panel- can you guys provide an alternative? Here is our requirements list based on what Condeco does”.

  • The must haves and nice to haves are immediately siloed to this one feature set: EG room booking. But how does it fit into a seamless UX with other elements of the building and workplace?  You are also just looking at what you have and asking how it could be a little better. Instead of discovering new opportunities to improve the UX or find new ways of working.

  • Network and security can be designed and scoped just like the UX. You should not rule out technology based on your assumptions from a POC. Instead you should work with the vendor or integrator to design the perfect balance of security and flexibility.

  • The scale of a POC is not an indication of real world performance. It is likely that products that scale well- find it hard to set up a small POC. And products that are aimed at small deployments do not scale well. And I've never seen a POC that is truely a production environment- so what is the value?

The IT selection process of running everything through a POC leads to selecting the wrong technology and 5 or 6 different technologies rather than a single integrated platform. The UX suffers greatly as each system has its own user interface.  This might be different systems with no integration across the following: visitor management, reception booking (room booking God-view, car space management, loan equipment and other requests), room booking panels, room finding maps, room availability reporting and room control. And this is just the rooms!

How do you scope UX across the building and tenancy?

The previous example shows how many siloed systems there can be in one tenancy basically just for rooms. But there’s even more silos to break through. These include kitchen, catering, desk finding/analytics, lockers, air quality (and other workplace wellness goals), retail and service integration, transport, media and digital signage, etc.. etc... . So how do we break through all these walls and help create a seamless UX? I think it is as simple as assigning someone to that role but first- scoping the UX effectively.

UX is not graphic design. The result of UX is your requirements list and IT's role is not to POC products- but to identify the integrations and dependencies needed to deliver the UX journey. You start the process by mapping all your types of users and building their personas. Why are they in the building? When/why are they not in the building? How do they meet/ collaborate What tasks do they perform?  How do they socialise? Do they live near the office? What are their fitness goals?

You then list all the spaces (assuming you have already gone through a process with Interior designer and know what spaces you need). Then for each persona you ask- what happens in this space… and this one when they are in it? And what is the journey from one space to another. Basically, how do they get to the building and what is every interaction with space throughout the day?

Finally, imagine everything is connected- what opportunities for workflow automation are there? i.e. can we replace the need for any user interaction and trigger actions automatically from data and/or profile information? Remove any double handling and aim for a single record or truth.

The main tip here is don’t just list every type of user and every type of space without thinking of the journey between spaces. Automate as many workflows as possible.

The UX scope will result in a long list of dependencies. At this stage it is important to have someone that sits across all the stakeholders to make sure the project can access all the required dependencies.  This will be across the three main areas of a smart workplace:

  1. Facilities / Real Estate

  2. IT

  3. People and culture.

Don’t get me wrong about my position on the  IT department. I’m not trying to undermine their value in the earlier sections of this post. I just think they need to focus on exactly that: their value. And that’s project related- not UX scoping. There is one project (my colleagues will know the one) that comes to mind where our client did not have an IT department. They are a real estate company so they tick box one and we had a number of stakeholders from HR so that’s box three. But two years later, and the job is still open. They had great ideas and scoped really cool user experiences but there was no one internally steering them to the right dependencies. And we were stuck trying to do something that was sometimes impossible because of fundamental IT problems.  

Some of our best work at ACA ,such as PwC, had all three areas above working closely together. If any one of those is missing from the mix it is, at the very least, a hard project to close off. And at worst- a failed project.

I’ll end with a list of easy to achieve workflow automation ideas if everything is connected to a single record of truth and a single automation platform.  But please let me know your own ideas!

  • Room booking: automatic workflow: Visitor management pre-population

If you invite external visitors via exchange/O365/Google/IBM Notes- then this should pass onto the visitor management system so reception can have a list of who’s coming today. This will speed up the sign-in process. We could of course automate a lot more but this is a good starting point.

  • Room booking: automatic workflow: Room Control

If you search for a room based on requirements- e.g. “I need a room with a TV,” the room should be able to automatically carry out a function related to those requirements like turning on the TV and initiating wireless sharing.

  • Room booking: automatic workflow: catering requests.

Most room catering in a workplace is simple- it shouldn’t require a seperate catering system. Map out all the possible catering rules and work out if you can just apply those as meta-data in your room booking. And if the room booking system is already tied to the email platform (exchange) you can use the service account to send requests to the kitchen.

  • Room Check In: automatic workflow: video conference dialling

Most conferencing end-points have great APIs. Including web-based platforms such as SfB. So if you have everything booked- the call should be up and running by the time both ends turn up to their space. Just walk in and start collaborating.

  • Visitor management: automatic workflow: car space allocation

Room booking, visitors and car space- could all be from one record of truth. Map your workflows between these three areas. Allocate car spaces, apply payment rules to different types of users (e.g. open the boom gate automatically for VIPs)

  • Visitor management: automatic workflow: building/floor/room access

If we go beyond just having a list of pre-populated visitors- we could email QR codes to external visitors and allow them temporary access to the building based around their booking.  Obviously consider your security options but we have allowed this for a number of projects.

  • Desk Finding: automatic workflow: Locker allocation and control

Find a desk- allocate a locker. Sounds simple but a lot of people run two seperate systems for this in a hot desking environment. Classic example of the problem I described in “UX is not about technology- so don’t let your IT department run the process. “

  • Desk Finding: automatic workflow: user preferences

Allow users to save their favourite preferences  to be triggered when they arrive at a desk or room. This could be desk height, lighting levels, climate and even favourite coffee order for different times of the day.

  • Locker booking: automatic workflow: laundry services

Maybe once a week you have it set to have laundry access your locker and take any washing to return at the end of the day. This can automatically charge your account and the system can provide temporary access to your locker- even if you change lockers every day. This could be used for any service. Basically a locker could be a personal PO box in your workplace.

  • Room booking: automatic workflow: transport timetable.

Most cities have a public API for transport timetable data. Could this be used in room booking rules- or to improve any other user experience?  

  • Coffee order: automatic workflow: payment processing

Have rules that decide when the company pays for catering (e.g. requires an external customer to be with the employee) and when to automatically charges any stored credit card the employee can add/manage.

  • Slack chat: automatic workflow: room booking and/or visitor management

If you use slack or similar you will find your employees almost live  there most of the day. So why make them use a different interface? Bots can allow for any feature provided by other user interface elements. e.g. “Hey @deskbot find me a space with a standing desk”

  • Friday Drinks: automatic workflow: vending machine control.

This is one we haven’t done. But is very easy with the right dependencies. We have been playing around with these new vending machines that can mix drinks and make cocktails. We have access to their control unit and can send commands from our triggers. Therefore- the input can be anything you can think of. A simple order page or how about this idea: our Chat Bot can recognise key words like “Friday” and “Drinks”. We can automate your Friday drinks booking in any room- and have your favourite drinks automatically dispensed from one of the smart vending machines. This is something we are going to set up at ACA. But vending machines in general are at a point where anyone could have their own Amazon Go like experience in their tenancy. Let me know if you have any ideas for smart vending machines in your workplace or smart building.


Takeaways;

The Problem

  • There is currently user interface overload in a building and tenancy

  • This is due to silos between the tenant and building. And additional silos at the tenancy such as the workplace.

  • This is amplified by technology purchasing and scoping coming from IT.

  • This creates a product first approach- missing out on new opportunities to improve the UX by perhaps removing the product altogether.

  • It also leads to more of the same- “let’s replace one room booking system with another” when the underlying problem is the UX not the technology providing the UX.

Recommendations

  • Have business units and HR scope workplace features

  • Consider the journey between spaces. If this is a new fit-out or building- the Interior Architects would be all over this.

  • Aim for workflow automation and single records of truth.

  • Have IT identify all the integrations and dependencies required to deliver the desired user experience


Jon McFarlane.

Dimension Data Sydney Office

Dimension Data uses ACAEngine for a number of workplace features. This includes desk usage and availability, people searching across a hot desk environment, room booking and space analytics. Most importantly, ACAEngine is a platform that enables these features by integrating into existing building and IT services. For example, to track available desks and find people- ACAEngine connects to Cisco wireless access points. These WAPs also provide the general network for staff meaning ACAEngine is leveraging something that is already there. No additional sensors, hardware or installation was required and we were able to provide better ROI on existing technology.


Notes on the growth of ACAEngine

This post was first published as a Linkedin Pulse article by Jon McFarlane. View other Pulse articles here.

Last month William Le from our team relocated from Sydney to Hong Kong to lead our efforts in the Asian market. In this short time we have delivered ACAEngine to a large insurance company for visitor management and building access control, started scoping our first major residential project and established a delivery and hosting partnership with two of the leaders in the Asian market.

Our aim now is to replicate this success in cities all over the world. Right now we have active projects in major cities such as London and New York- but also a few scattered in smaller cities such as Milan, Istanbul and Ho Chi Minh. This is thanks to some of our multinational clients- managing the solution centrally and rolling it out to their global offices.

How do we manage this with a relatively small team? Well there are a few things working in our favour.

Scalable product from day one.

Our software was designed to scale from the get go. Many startups are following the “lean” methodology- basically releasing the Minimum Viable Product (MVP) and constantly playing catch up. This means they focus on features that they can sell right away. Our first project was a single system- it would have been easy for us to ignore scale as the client didn’t need it. But we took the time (while living on 2 min noodles) to get it right before releasing. Having the technical co-founder that I do, he would not of had it any other way. Eight years later, scale is still at the heart of what we do. We are a couple of months away from a new version that can scale even further. We even built our own open-source framework so we are not held back by Ruby on Rails limitations. Our current offering is as scalable as our clients need right now, so again we are ignoring the MVP method and releasing something that will be scalable for another eight years- while our competitors are yet to catch up to our day one.


Software has no borders

The building industry has historically been very hardware focused when it comes to technology. As we are 100% software based, international business is pretty straight forward. Even when we work on projects in our own city- we are doing it remotely. From a technical point of view we do not care where the server is located, the process is the same.

Instagram famously had just 13 staff when they were acquired for $1B. We are not running that lean, but it shows how a small team can have big results when all they focus on is software. This is one of the main driving factors behind our internal shift of delivering our projects directly; to working through partners so we can focus on product and not spend half our day in project meetings.

Furthermore, many of our elements and tools are open source. This means we have external developers contributing to some elements of our stack. And any partner or developer can get up to speed very quickly with our documented APIs.

The open source model also allows clients to set up their own demo or Proof of Concept as they can download and setup ACAEngine without us. If you are not using it commercially, you do not need a commercial license. This works really well for us as resourcing POCs is hard even for the largest IT companies. Also, not needing us helps our clients validate the platform so this turns out to be a successful way to run a POC.

Partnerships

Projects require local resources- so we lean on partnerships to enable this. There are two core areas for our delivery partnerships:

  1. Designing, scoping and building UX

  2. IT services and integration.

As an API based platform, ACAEngine UX can be completely customised for the job. Companies that typically sell product can struggle to get their head around this. So we also provide templates for a number typical requirements such as a workplace app, room booking interfaces and visitor kiosk (to name a few).

Creative agencies are best suited to scope and design UX as they already work with and understand platform. Our message is not too far off their typical web projects: “We’re Wordpress and you're engaged to build the website on top” is the simplified analogy. "Wordpress for buildings".   

IT services are simple: we need a VM (maybe a few), client side Auth, a network and domains with SSL. Our challenge is these two areas, UX and IT, might not be covered by the same partner. Personally, I spend a lot of my time travelling and meeting with potential partners to close these gaps so the client can engage one partner and have both areas covered. 

So what next for ACAEngine?

We are going to establish more partnerships- not going to go nuts with it rather be selective and focus on capabilities. We are launching new products Q2 this year (a SaaS offering and our own building Analytics platform) and we are using our existing projects in new markets as a base to grow sales in those regions. It will be a busy year for us- but I guess it has never been quiet. It’s always just a new set of challenges- but we have a motivated team that thrive on it.

It you are interested in my personal experience as we do all this I've started a personal blog here: http://www.jon.sydney/

Jon McFarlane, co-founder ACA



Apps- the new 'box' you don't need.

This post was first published as a Linkedin Pulse article by Jon McFarlane. View other Pulse articles here.

apps image.jpeg

An app can do a set number of things. If you want it to do anything more-  the developers have to build and push a new version and users have to download an update. The company that develops the App is not set up to scope UX specifically for you.  This might be fine for the average consumer but is not suitable for a smart building or smart workplace project.

I feel like we have been here before in the integration industry. It reminds me of the rack full of boxes that have a set number of features and are sold to the client with little consideration for the total solution.  But instead of 5 boxes per room- the industry has switched to 5 Apps. And most of the time, these Apps are to help sell more boxes!

The integration, AV and IoT industry is so hardware focused that their answer to software is mobile Apps- in 2019! A clear example of how far behind the broader IT industry these fields are. And how disconnected from the customers they have become.   

There are countless articles on the App being dead but here are a few:

The software industry is focused on platform, APIs and build your own UX. The UX is not just button presses in an App, although this can be part of it and many of these platforms have their own App. But customers are thinking more broadly and aiming for passive interaction with technology.  The UX might be automatic triggers from user location, predictive actions based on historical data, chat-bot integration into Slack or a new type of interaction we are not thinking of yet. With a platform, you have this flexibility.

Apps and boxes are easier to sell- as it requires less of the sales person. This is how they end up in so many “smart office” projects. Another room booking app- another room booking panel. Not very smart.  This is why I wrote an article on not attending industry events. I learn more from my customers, partners and co-workers than room booking app vendors invited to talk about the future of workplace.

Think about some of the apps in this field; most require their hardware. “Our App tracks people for automatic room check-in. But you have to buy our bluetooth beacons”.  Or“you have to buy our desk sensor”. Or “you have to buy our in room control hardware”. Or “you have to buy our video conference system”. Or “you have to buy our swipe card readers”… etc, etc. And if they even have the ability to integrate outside of their App- you probably have to buy their middleware software or subscribe to their public cloud.

The customers are demanding more. They don’t want all their building technology to be disconnected systems that required multiple apps to operate. The end user certainly does not want an app for the car parking, an app for their visitor management check-in, an app for their elevator, an app for their virtual access card, an app for their locker, an app for the desk finding and an app for their room booking. From the articles links above- you will see that App downloads per month for an average user is almost 0. When was the last time you downloaded an App that you didn't delete a few hours later?

My final thoughts are for system integrators in this field. IT companies and creative agencies have been working with platforms and solution-based sales for a long time. If you don’t step up- they will take over your AV, IoT and integration niche. It’s only a niche because it required special hardware and weird vendor specific software. It doesn’t anymore. You might be a billion dollar system integrator- but your vendors are blinding you to the opportunities and changes in the market. Meanwhile,  companies you have never heard of are taking your market share. Stop selling boxes, stop selling apps and focus on solutions.


Jon McFarlane. Co-Founder, ACA.

Introducing Spider-Gazelle- our open source web framework

By Stephen Von Takach

Since the founding of ACA Projects the I.T. industry has gone through a number of rapid changes. The rise of cloud computing, followed by new ways of deploying and managing applications has meant almost everything  has changed.

  • 1991 First web browser

  • 1996 Ruby released… (NOTE:: it takes a few years before adoption of each technology is widespread)

  • 2004 Network stacks: Linux introduces epoll

  • 2010 Service managers: init is replaced by systemd

  • 2013 Application deployment: packages replaced by containers (such as docker)

  • Server management: Puppet (2005) -> Chef (2009) -> Kubernetes (2015)

One advantage of these changes is that it is now easier than ever to deploy services, making microservices a highly attractive option for implementing new features.

A NEW FRAMEWORK

Since 2012, ACA Projects has been almost entirely a Ruby on Rails house however we’ve been working on a framework that makes it easier for us to rapidly build and deploy new services.

One of the most exciting things is that it is fast, in fact it’s one of the fastest frameworks in the world: 

111111.png

For those paying attention, that’s 3 times faster than NodeJS and 32 times faster than Rails.

222222.png

 

NOT ONLY FAST

Coming from Ruby, Crystal lang was the obvious choice for the underlying language of the framework. Effectively “compiled ruby” it is a joy to work with when compared to languages like Go or Rust which are much more akin to modern versions of Java or C++.

If you’re a developer coming from Rails, you’ll instantly feel at home with Spider-Gazelle as one of our major goals was to ensure a seamless mapping of skills. Rails is an extremely feature rich framework, it makes your life as a developer easy and we didn’t want to compromise developer productivity or happiness. As such, we have ported most of the core Rails features to Spider-Gazelle without compromising elegance or speed.

From a deployment standpoint, Spider-Gazelle projects generate deployment images approximately 45MB in size compared to 850MB for Rails, 18 times smaller.

 

PRODUCTION READY

We have a few Spider-Gazelle projects running in production. Meraki Scanner, for example, is running at multiple client sites and processes wireless location data streamed from Meraki wireless access points - tracking people and devices in buildings.

OPEN TO ALL

We believe in the open source movement and as with many of our endeavours, are proud to give back to the community.

For anyone interested in learning more about this exciting new project, please have a look at our documentation: https://spider-gazelle.net/