Giffgaff sells its soul to AWS
UK MVNO Giffgaff says it has gone ‘all-in’ on Amazon Web Services to power all of its technology infrastructure and application development.
As ever when a company decides to entrust its entire technological fortunes to a third party in the form of a move to the public cloud, a boilerplate list of features and benefits has been employed. “To improve its members’ experience and accelerate innovation and development, Giffgaff is leveraging AWS’s proven scalability, security, agility, and performance.” Leveraging, ugh.
Giffgaff says it’s going to put data at the heart of its business, who’da thought it? Apparently this can only be done by shifting all its databases and applications to servers it rents from someone else. To be fair, AWS does claim to add a lot of value once you make that move and Giffgaff is going to use at least 60 of its extras, including compute, analytics, storage, databases, containers, and machine learning.
“We started out with a traditional, on-premises infrastructure, but the need for ongoing maintenance made this model overwhelming for our technical team,” said Steve MacDonald, COO and CTO at Giffgaff. “For example, it used to take us up to two weeks to provision a new server.
“When we began to adopt AWS, we were able to turbocharge our development lifecycle by focusing on innovation rather than wasting time on maintenance. It’s such a powerful capability for a digital-native business like ours. The process of constantly evolving our software means that we can always deliver an award-winning service to our members. Our plan now is to extend the use of AWS, and make our members’ voices serve as one of the primary data sources for business decision making.”
Sounds like your team was a bit rubbish then, Steve, and if so whose fault was that? The primary justification emerging from the above quote seems to be less about agility, scalability, etc, and more about saving money and outsourcing technical stuff to a more competent third party, as is so often the case.
“Giffgaff is demonstrating its ambition to be a data-driven company that can quickly scale and innovate to meet the needs of its members,” said Darren Hardman, GM, UK & Ireland at AWS. “It’s exciting to see how Giffgaff is using cloud technology to transform its business – and the industry – and we look forward to working alongside them as they leverage the breadth and depth of AWS to drive innovation and a compelling customer experience.”
It boggles the mind to think about how much lovely leveraging Giffgaff is going to be able to do now. Of course, now that it’s umbilically dependent on AWS for everything, the latter is free to leverage its prices and terms of business in whatever direction it wants in future. And if Giffgaff doesn’t like the changes, it now has the agility to leverage itself out of business.
What’s with the snotty attitude from the 20th Century? Its likely we’ll see more of this including from large established players. If it drives down costs and reduces customer fees, its welcome.
So just accept it as inevitable then and don’t ask any questions?
I’ve installed messaging systems SMS MMS voicemail etc at UK and international telcos getting hardware installed connected and commissioned can take 6 months you miscalculate the capacity. even pulling out all the stops it took 4 weeks to put a new server in at vha to an existing rack. Aws I can spin up 6 servers in 30 minutes and process 20 million SMS for spam an virus control load testing in 40 minutes then delete them Again in 10 minutes so in half a day it took what would have cost 50k and 6 months to do previously and I’m billed about 6 quid no brainer really. And all without the soul sapping meetings on meetings and the network guy getting involved with the infrastructure guy and the operations guy and then having to get budget approval from the investment review board and then having 2 project managers etc etc.
This was so badly writted I thought it was a satire piece.
Badly writted – love it.
I thought it was well writed ???
Many companies -including telcos – are moving workloads to the cloud. And many more are picking one cloud as their primary cloud. Having worked with AWS, Microsoft, Google, IBM, Alibaba cloud for a few years, it is not surprising to see a company move to a single cloud – to focus their teams skillsets in one area. Not quite sure what your point is here Scott – are you surprised about this move, critical of it, would you prefer another cloud?
Just sceptical about this unconditional acceptance of what a great idea it is to relinquish so much control.
This is a poorly written article, full of unnecessary attitude and uninformed / out of date views. Moving to the Cloud has been the accepted path for a majority of enterprises for circa. 3 – 4 years. If anything, GiffGaff are late to the game!
The accepted path eh? Well I stand corrected then John.
Wow, such a poor uninformed piece of journalism…
As others have mentioned this comes across as a poorly researched and written opinion piece on a personal blog rather than anything professional.
Your points and concerns may be valid, but qualifying them by calling GiffGaffs technical teams useless and simply hating on cloud providers with no real justification makes this nothing more than a rant.
Having worked in physical infrastructure I can attest that no matter how good your engineers are, getting a server ordered and shipped to a data center and racked and networked and configured can easily take weeks without either major forward planning, overpurchasing, or both. Cloud providers allow much faster scaling up and down at a fraction of cost. This could be a smart move on GiffGaffs part, but this article seems ignorant to the benefits and downsides of physically maintaining your own infrastructure.
“calling GiffGaffs technical teams useless” I didn’t.
I liked this piece. Scott is correct about surrendering to the behemoth that is Amazon. In ten years time there will not be any option for on-premise if everyone continues down this path. At that point the price of services will be uncontrollable, and the choice generally will be close to zero. Cloud brings lots of benefits, as does buying stuff from Amazon’s marketplace. The long term consequences of both are monopoly supply, and rather scarily, monumental amounts of un-policable harvested data. The future’s bright…
You’ve probably missed the point a little. It’s not about giving up control because your staff aren’t good enough, it’s about focusing your time, effort and ultimately money into areas that set you as a business apart. Are giff gaff in the hosting business? no, they are in the telecoms service provider business. For them the former is simply a facilitator to the latter. If that can be farmed out to someone else to make it, cheaper and more reliable with less effort, it frees up staff to concentrate on improving their offering to customers and hence increasing the money that comes in. Sure transitioning to the cloud isn’t the simple process everyone pretends it is, but if done right the benefits are huge
Sure, but where does the outsourcing of non-core competencies stop and when do we start questioning the downsides of this faustian pact?
I agree with the other comments, what is this article talking about? Moving things to the cloud isn’t a bad thing, and if that one provider raises then they’re free to move to other clouds. They planned a migration from in house to cloud, they can do the same from one cloud provider to another.
Not a pro on the mobile world but isn’t Giffgaff completely dependant on the Three network, isn’t that worse?
This should be taken as good news and hopefully you’ll learn and see the light.
“See the light”. LOL, I hadn’t realised how zealous public cloud advocates are.
I wonder how easy it actually is to leave the Hotel Cloudifornia.
Yes, it’s an MVNO, that’s the whole business model mate.
You clearly have no idea what you’re talking about. You obviously have a fundamental misunderstanding of the following:
1. What it takes to operate and maintain a modern, fault-tolerant, multi-site, multi-region datacentre.
2. The software required in order to make the hardware accessible to the application developers (this was alluded to in the “two weeks for a VM” comment).
3. What cloud services actually offer: security, reliability, agility and so much more. These things directly effect how quickly and successfully companies are able to innovate.
I would have thought that a technology journalist would have been able to figure out that running a datacentre isn’t one of giffgaff’s core competencies. Much larger and wealthier FTSE100 companies have tried and failed in this aspect (source: I’ve worked in a number of them and they’re shifting to cloud as fast as they can, where appropriate).
Stop knocking other comments, educate yourself in this area and swap the snark for some humility.
And if I don’t?
Finally a bold article asking questions about blind faith and move to cloud. “it used to take us up to two weeks to provision a new server”…??. Because we couldn’t build the servers fast enough we decided to pay 5-7x more for the same compute and as a tech company lock the future with one company.. amazing!!
I wish people understand that GAM (Google AWS Microsoft) the Trillion+ market cap companies are killing innovation. What supermarkets did to small retailers is finally happening in the software industry.
You replied to a previous comment implying that we shouldn’t just accept it, but instead question things like this. There were no questions posed here, except a sarcastic “who’da thought it?”.
So what questions are you posing here exactly? What is actually your point of this article? I challenge you to not reply sarcastically, or make fun of my grammar, which has been the approach for all other responses here.
Read all of the comments.
“Something went wrong” https://www.bbc.co.uk/news/technology-55299779
This all just reads like the words of a completely uniformed person who also has to be very bitter and dismissive on top of that. If you had any tech or even business strategy knowledge you would understand why cloud makes far more sense than on-premise.
This whole idea of surrendering control is also without any real basis as someone can easily switch to another cloud provider like Google or Microsoft. It’s normal to start with one cloud provider and then explore others. Or now you’re going to come and tell me that all of these cloud providers are in some sort of dark secret club and how they will plot together to take control? If so ignore my other points as you’re far beyond having your opinion changed.
Whatever you say, Dick.
In the industrial revolution many (initially most) companies generated their own power on site. By now, almost all have settled on taking utility grid power. Did it make us all dependent on the electric grid operators? Sure. Are there power cuts? Sometimes. But can you cost effectively run a major facility off grid and out compete your rivals? Probably not.
This is all that’s happening with public cloud. For all the downsides, it’s compelling.
By that logic, the cloud players should be regulated as utilities.
Scott…with respect, you are fundamentally lacking in understanding around what AWS is and what organisations use it for and your article evidences this.
There comes a point where you need to look at the comments and realise that you’re talking absolute garbage. AWS is a $35bn/year business (and growing by ~30% YoY) for a reason and it isn’t because of outsourcing/relinquishing of control…and it isn’t because of some inherent lack of skill with in-house IT.
Utterly ill-informed drivel.
Your argument: AWS is making loads of money so it must be good and we should unquestioningly relinquish control to it.
On what planet do people relinquish control to AWS by taking their services? I’d argue the opposite…the economies of scale, the ability to template architecture, the freedom to innovate and “fail fast” with zero capital expenditure and near limitless capacity, potential and geographic reach allows customers to focus on customers, services and everything that is “in the cloud” as opposed to wasting time on maintaining architecture or worrying about capacity planning/infrastructure lifecycle.
Netflix, Spotify, Uber, Zoom…none of these would have been possible or at least would have been severely delayed, costly or stunted without Public Cloud.
You clearly have no experience of Enterprise Architecture and you clearly have no real experience of Public Cloud…as such, this reads as ill-informed, keyboard-warrior drivel.
Oh Dean, you were doing so well, but you just couldn’t resist lapsing into ad verecundiam and ad hominem, could you? You clearly have no experience of Logical Fallacies and you clearly have no real experience of Rigorous Thought…as such, this reads as ill-informed, entry-level drivel.
It’s a fair opinion but what do you suggest as an alternative? Clearly they don’t have the capability to do on prem and there are of course options like Google cloud and azure if aws gets too expensive. Massive cloud companies are not evil, they’re providing a competitive service and there does exist enough competition to stop a monopoly.
I don’t suggest an alternative, I’m just questioning how wise it is to relinquish so much control to an oligopoly.
You do realise that moving ro AWS is no more than renting someone else’s hardware, right? And that someone needs to take care of the hardware, power, network, so you can focus on developing your apps. At any point you can migrate to another cloud provider, or to on site server. After all, what you essentially use AWS for is to send your code there to be run, but any responsibile developer tests it before it is deployed, so basically there would be a local environment that runs it anyways, which can be turned into a server.
AWS and other cloud providers guarantee amazingly good up time, response etc. all that for a fraction of a cost that you would have to spend on your own infrastructure and people to maintain it. When AWS ec2 goes down, it falls back to another instance in selected by yourself location, the downtome would be minimal of not none. With your own infrastructure, you will need to get someone to handle it, and if that happens over the night, you may have several hours of downtome before it is even noticed…
AWS does jot control your data or software, you are free to take ot elsewhere or destroy it, they only provide you with the hardware. By the logic of your argument, we should not use operating systems either, as these truly control your data and may damage hardware you paid for, furthermore we should not use emails, messangers, websites etc… The article you written is controlled by someone else, and most likely hosted in the cloud.
My only argument is that going ‘all-in’ in the public cloud is not unconditionally positive.
do you actually believe your own opinion or do you just relish the attention of being controversial? The frequency of your comment responses would suggest the latter
False dichotomy.
Scott, you clearly have a problem with cloud services, but also clearly do not understand them. The cloud offers serious advantages for businesses, and those not embracing the cloud are slower to react and will get left behind. It is not true to say once you are ‘all in’ in the cloud you are beholden to the CSP. Services can be moved between CSPs or back in house if designed correctly.
Your snarky comments in the article and in your replies come across as incredibly unprofessional and childish.
I suggest if you want to slag something off, and the people who clearly know more than you, that you do your research.
Clearly.
There is a point about Cloud Providers being regulated as utilities and that will have to come in some way, probably in ways that make it easy to migrate from one to the other or mix and match different services from different providers.
But outsourcing to the cloud is a game changer, not just in managing infrastructure but also for software development. As an example, how would you build microservices that you could spin up and shut down within a second if you were working on prem? With near infinite scalability? Not saying you can’t, but it’s a lot easier on cloud.
What if cloud goes wrong like yesterday’s Google outage with their IAM service? Major issue resolved within an hour. I’d be chuffed if I managed that on prem.
Cloud carries risks, the worst of which is lock in. For most organisations though it lets them build and run applications on scalable resilient infrastructure that would otherwise require shed loads of kit and people.
Lots of good points there Ken and no hysterical ad hominems. Thanks.
the best comment on this piece tbqh
Well this piece was a total waste of space. And what’s with the attitude? Not very professional, was it?
It didn’t take up any space.
Wow here I was thinking trolls just wrote comments, now they’re writing the articles too.
These companies are always going to be dependant on somebody for their infrastructure even if it’s internal. And it’s perfectly normal for companies to have suppliers when they grow too large to provide the service internally.
This is no different to them having Dell provide several thousands of work stations instead of having a team dedicated to building PCs from scratch.
Of course there are risks involved when you put all your eggs in one basket but they likely out weigh the risks of trying to scale with their internal team.
It seems this is just another attempt at a controversial opinion piece with a clickbait headline to try and drive some traffic. I’m ashamed to say it worked
Thank god for ad block at least
You’re only human.
Well if it took your engineers that long to provision a new server… Sounds like two things you need new engineers and that the UK is known for underpaying all of the IT field. I compared my US salary with the UK equivalent position and if I was working on the UK market instead of the US I would be making 75% less. Which is why you get 2 week deployment times….
Do us a solid and Please change your career.
Do us a solid?
It’s been fun sparring with you crazy kids, but I’m on holiday now so I’m calling it a day. I’ll leave you with the thought that there is one cloud service I’m really enjoying right now, thanks to this story, and that’s Google Analytics. Have a good Christmas everyone.