Are integration and automation always the right approach?
By Jackie Green, CTO, Alteros
Workflow. Compatibility. Integration. Automation. Efficiency. The business world has become an absolute sucker for these words. We like the idea of things seamlessly working together; a vision of uniformly branded hardware, intuitive GUI, a single, smooth, seamless system that allows us to feel that we have time to get on with the serious issues of the day – ‘bluesky thinking’ (!?), the creative process, the stuff that matters – without being pulled down by the daily frustration of technical logistics, of simply ‘making things work’.
The desire for this is both intuitive and strong. How many times have you encountered the following situation: Tom from Marketing wants to send you a report he’s just written. You intend to review it, make a couple of changes, and send it on to Laura, your manager. Tom uses Pages on his Mac, you can only find your old laptop that still seems to be running Word 2003, and Laura is a big fan of Google Docs. You’ll probably need to end up putting the document online as a PDF too.
It’s no big deal really. Some of these things are compatible with each other. Word 2003 gives you some hassle but you can quickly google an online converter and sort it out. And yeah, a few of the tables Tom created are now a bit wonky – and the font that’s part of the brand’s accepted style guide has gone missing. But hey-ho, that’s how it goes, no?
Except we’re recognising that exactly this kind of situation is the ultimate productivity drain. Bottlenecks in the process that add not only unnecessary minutes to the day, but unnecessary psychological obstacles: if you’ve been dreading working on something, any obstacle to getting started can be the difference between it getting done, or you going and making a 5th cup of coffee and staring out of the window for an hour.
Imagine the alternative though. You get Tom, Laura and the whole rest of your team in the room and you tell them ‘right, enough is enough, none of this document-opening malarkey anymore, we’re all going to start using the same software’. Now you need to have a two hour meeting (initially scheduled as just half an hour) to decide what’s best to transfer to. Nobody can decide whether Cloud storage is a great idea or a terrible security risk. Finance is skeptical about shelling out for licenses when there are open source options available. Bev swears by Apple products and won’t use anything else, whilst Robert, has used Word for 20 years, and you all know he will take literally years to get his head round something as simple as saving a document in whatever new application you decide upon.
Even though everybody now has files that are compatible with each other, there’s a lack of productivity as everybody tries to accommodate the new approach, and – worse – a huge amount of animosity and tension. The apparently easy and obvious decision to harmonise, integrate, automate – it has suddenly become a seriously big headache.
I’ve deliberately avoided using an audio hardware or software example here, since not all my followers are necessarily directly from the audio side of things, and because the ‘problem’ of productivity is absolutely universal. But of course if you do work in audio you can instantly conjure up your own version of the same problem.
I set about this piece with a hope of overviewing the academic business and engineering literature on the subject – identifying whether there were any quantitative stats that could add a bit of heft to either side of the argument: to automate, to integrate – or not. The first thing I found is that there’s a serious dearth of literature on the issue, at least in relation to methodologically robust, independent research. Of course there’s the standard abundance of marketing messages from companies peddling various integration systems; business process reengineering, media management systems – you name it, there’s a business aiming to streamline your workflow.
In terms of academic literature though, it’s really surprising to find so little. There are some apparent obstacles; firstly, it’s a hard issue to explore in the general sense. This means there’s quite a bit of literature examining workflow efficiencies through systems integration in specific fields (healthcare is a particular one), but ultimately this specificity indicates there’s a fair amount of contextual dependence, with some fields better suited to automation and systems integration than others.
The more general studies that exist – such as that conducted by Bakar – are quick to identify benefits from system integration, but there’s very little indicating the drawbacks. Is this because system integration is so good that there really aren’t any drawbacks? Does it always improve workflow efficiency? With the lack of quantitative evidence addressing the question, my assessment of the issue has to be based somewhat on personal interpretation, a little bit of anecdotal evidence, and some ‘softer’ literature – but answers to the key questions can none-the-less be found.
The first of these is – what modifies the overall ‘integration is great’ message that permeates most of the commercial online messages? HISCOX identifies that whether you’re working in real-time or not can make a significant difference to whether you want full integration, or whether you’ll simply accept interfacing to make things talk to each other properly. Interfacing adds another layer to data processing which may add a delay to when these elements are ‘prompted’ to talk to each other or the central bridge that allows them to communicate with each other. In the pro audio world, those bridges and switches play an important role in the final quality and speed of the output.
Security is also a potential concern: with an integrated system, access to one component of the chain means control over the full system. Deciding to integrate means thinking very carefully about security and access. Do you have access to dedicated resources such as fiber or satellite? Integrating systems from coast to coast can make production incredibly efficient and cost-effective, but it wouldn’t be practical for a user just going through their typical internet access provider.
Another point to consider is where you are in the process of developing your systems approach. If you have legacy systems and everybody within the organisation is very happy with how they work, then a changeover to a singular, integrated system may be both complex and expensive. Your assembly of hardware and software up to this point may have been piecemeal, before future-proofing and forwards integration were key concerns, and thus your focus in new acquisitions will simply be on finding solutions that have the flexibility to work with many different legacy systems as effectively as possible.
But if you’re about to embark on a completely new project and need both the hardware and software to get going, then certainly your life is likely to be made easier if you look for integrated systems; lower complexity during installation and operation, fewer technical details for your staff to get to grips with as they learn their way around new interfaces and modes of operation, fewer steps in the workflow, more production flexibility and possibly either less staff to operate or more general staff rather than specialist trained. The only exception to this will be if you have very specific technical specifications that can’t be met by the ‘one-size-fits-all’ nature that often comes with fully integrated, holistic solutions. For these, it may be necessary to pick up individual components, and then find a way to make them speak to each other. To that end, I have recently seen a rise of in-house development resources to address these very unique cases.
Integrated systems that cover the full engineering or business process can be more beneficial for engaging in diagnostics and monitoring, giving a full 360* of how the system is functioning, whether it’s delivering the results you want, and where slowdowns or problems in the system are occurring. Again, I’m not talking in specific terms relating to audio production and engineering, because these really are general messages that apply across workflows in practically any industry.
What about the disadvantages though? Well, I already mentioned that in highly technical cases with specific needs, integrated systems may not hold all the answers, and lack the flexibility to do exactly what you want, how you want and when you want. Integrated systems may mean adjusting your processes to meet what the integrated product can deliver, rather than finding something that perfectly fits your existing operational context. That will work for some, but not for others.
Then there’s the issue of cost. Many integrated solution providers – regardless of the market – are quick to say that using an integrated solution is cheaper. Often this will be the case, at least initially: the kind of value that comes from ‘package-buy’ discounts and the fact that single pieces of equipment might be able to achieve multiple functions. But it is important to remember that starting off with integrated solutions is going to tie you in the long term, meaning that the ability to shop around in the future might be limited. This is where standards such as SMPTE2110 can play an important role in ensuring more flexibility with future options.
It would of course be remiss of me at this point to not talk about the approach we’ve decided to take at Alteros. I think it’s fair to say we count as an integrated system. Traditional audio setups for events tend to include the use of custom cavity filters and tuned antennas, along with custom distribution amplifiers and combiners. These are expensive (to purchase, but even to rent), they’re complex to setup, and trying to get these things to coordinate effectively is fraught with potential problems. By eliminating this we really do think that we’ve capitalised on the benefits that you get from integration: ease of use – even by non-technical experts – is embedded right into the system we provide.
That said, we‘ve also aimed to make our systems work in a way that doesn’t mean you’re exclusively tied to our provisions and nothing else; allowing the use of fiber or low-cost Cat 5 cable to link our systems was a key consideration as we developed the system. We didn’t go for building a single ‘one-size-fits-all, jack of all trades’ system, but instead a component system that gives you the building blocks to tailor a solution that meets your needs and is scalable as your audio production needs change – and with up to 64 transceivers available for operation in the system, there’s few spaces or workflows that we can’t match. All of our software solutions: the transmitter screen, the audio group screen and the settings screen, aim to give that 360 holistic overview that I talked about, and to make coordination of the various elements largely seamless and intuitive.
Finally, we’ve got the direct-to-fibre extension that we launched in 2018. I think this more than anything demonstrates how we create an integrated system with the aim of maximising not only ease-of-use, but also keeping in mind the vital need to ‘future proof’ and maintain as much flexibility for the user as possible. The first benefit is the efficiency gains that are made from this integrative approach: IP provides incredible workflow efficiency, and when you eliminate the need for our system to engage in any conversion to make use of it, then you absolute maximise the efficiency and flexibility that you can get from an integrated system.
The second thing though is the quality. A natively digital system capitalising on direct-to-fibre integration eliminates possible jitter, resolution degradation, signal loss, increased latency, and generally negative effects on the sound quality. This is its own efficiency gain – not one achieved by a reduction in manpower or effort, but an efficiency that is inherent in the system – one where efficiency means the maximisation of effectiveness for the minimum of human effort. That’s what the GTX direct-to-fibre facilitates.
I’m not attempting to use this blog as an opportunity to tell you how wonderful the integrated nature of the GTX is (though it’s a helpful side-effect, sure). I genuinely think there’s benefit in having a critical debate on whether integration and automation are always and universally the right approach for every application – audio-based or otherwise. As with so many things, it’s a cost-benefit analysis, and we’ve attempted to pitch the Alteros product at a level that capitalises on the benefits whilst minimising the costs. If you have had opportunity to read any academic literature on the matter, I’d be fascinated to read these sources and hear your opinions.