Welcome to the Fasthosts ProActive Podcast: Spill the IT. Each episode, we'll sit down with some of the amazing ProActive team and chat through their experiences of the ups and downs of IT infrastructure management in small businesses. There's always plenty to chat about.

Infrastructure metrics - the discussion continues about data analysis and using your MSP's skills to drive business growth (and Dan talks about a powerful onion).

Listen on your favourite platform!
Want to listen on your go-to platform? We're on those too...

Episode transcript:

Intro (00:05):

Welcome to the Fasthosts ProActive Podcast: Spill the IT. Each episode, we'll sit down with some of the amazing ProActive team and chat through their experiences of the ups and downs of IT infrastructure management in small businesses. There's always plenty to chat about.

Graham (00:29):

We was talking earlier in a previous podcast, weren't we, about the Amazon experience, about retailers, and what we as humans, how we interact with the internet, when we're ordering something, what do we expect? Do you see, with customers, what they expect to get out of data, is there starting to form a model of what they're looking to get out of that?

Dan (00:48):

I think it depends where the customer is on the journey. So I think for those customers that haven't been doing anything in the kind of metrics and observability space previously, their first reaction is, "Wow, I've got some data here, I've got a lot of data here. What can this tell me?" And then, with that type of customer, it's guiding them through not becoming overwhelmed and guiding them in the right direction. I think some of the clients come to us much better informed, they're much more kind of mature in that journey and have very specific things that they're interested in.


I'm going to mention the Onion model at this point, all right? There's always a good kind of food analogy for these things. On the surface, you can understand about the platform that you're running on, the tin, what's it doing at any particular point. And as you start peeling those layers off, as a business, you should care less about how much compute, how much network, how much storage, all the rest of it, because we as a managed service provider take care of that, they should be caring more about, ultimately, what their customers are doing and how that should drive their business. So the more layers you peel back, the closer you get to understanding how you need to shape your business, rather than how you need to manage your infrastructure.

Graham (02:06):

So I guess that's all around tailoring, isn't it? At the end of the day, is how you are tailoring that information, how you collect that data, and what you do with it?

Gary (02:13):

And that's a skill. We've spoken on a number of the podcasts around skill shortages, and data interpretation is a skill and quite an expensive skill to procure. So one of the things that we can bring is that context and that interpretation that, A, we have individuals within the business and teams within the business whose job it is to interpret data because hey, we need it for our own clients and for our own business. And so then, we can share that expertise and that experience with our clients so that we can help them on that journey.


As Dan said, if it's something they've never done before, we can give them pointers and say, "Well, this is what we've done in the past." We're like chefs, we're eating our own cooking. So we've implemented these things because we've had to implement them for ourselves to consume as a business. We can now help you as a client of our business implement these things. And this is where, we've made the mistakes, we've learned from them. And now you get the benefit of that experience and the skills needed to run with that.

Graham (03:16):

Claire, from a human perspective, and previously, on our last podcast when we were talking about security, we were talking about the impact on people and their working conditions and things like that, from a success perspective, on a continued sort of customer success basis, what's the impact on people about getting the right data, assessing that data, and making informed decisions to then progress or to do the right thing in their business?

Gary (03:39):

It comes down to transparency. So one of the key pillars of the business is that we want to be transparent and by providing clients with the metrics that they want, and also perhaps some of the ones that they aren't sure that they want, but we're being completely transparent, this is exactly how your solution's performing.


We kind of avoid that finger pointing because we're not trying to hide anything. If we have an issue that is as a result of our own supporting platform, one of the first things we're doing is telling you about it. So we're not trying to hide anything. We are making everything clear, everything open. And as Claire's mentioned, that that builds trust. And when you're in a position with a client where you are both trusting each other to do the right thing for the business outputs that you need, finger pointing isn't needed. So.

Graham (04:39):

Yeah, because it is what it is. Yeah. Yeah.

Gary (04:40):

It is what it is and this is how we move forward from it, this is what we do with this information and this is the journey we take together as a provider and a consumer of those services to make things better.

Claire (04:53):

Hopefully, building then that trust with your client that they take your advice, they're listening to you, and that's all about building the relationship with the customer.

Graham (05:01):

So that sort of leads me on to my next question. Do people sort of focus on the wrong data across their applications? And how is this different in, say, a cloud environment versus to what they're looking at in an on-prem environment?

Gary (05:14):

I think the focuses are different out of necessity. So when you're operating on an on-prem environment, you don't always have the same agility and scalability that you do in the cloud. So there are certain data points that you absolutely have to focus on, which just become a non-issue in the cloud.


And I think a good example is some of the more traditional bare metal, for want of a better phrase, metrics, so disc space utilisation over time is a big, big thing with on-premise solutions because you've got those physical discs, if you run out of capacity, if you want more, you have to buy those discs. And we've already spoken in previous podcasts about hardware availability challenges. You then got to get somebody to install them, you've got to provision them, you've got all this stuff to do. So tracking that use over time is really, really critical.


In a cloud environment, it's a bit meh, doesn't really matter because if you suddenly notice you are sailing a bit close to the wind on your disc space, the process to remedy that is to raise a request to say, "Can I have more disc space?" and us as a provider saying, "Yes, there you go. There's your extra disc space." So it changes the focus and becomes less about consumption of, effectively, what's a commodity, more around focusing on data points that give you that proper insight. So yeah, absolutely different focus.

Dan (06:47):

I think the cloud also gives you additional data points when you talk about scalability and being able to grow a solution, maybe you're doing that for peaks. It's not also about ramping up ready for the peak, it's being able to know when to ramp down as well.

Graham (07:07):

Yeah, interesting. Yeah.

Dan (07:09):

Because you could easily ramp up and keep your provision at that level and that's going to cost you and then, you're no better than being kind of on-prem really because you bought a static resource level. So I think there are constantly introduced other data points that need to be observed as well, which just wouldn't be necessary like Gary said in a kind of an on-prem solution.

Graham (07:30):

So in essence, are people sort of lazy when it comes to data? Do they not want to do the heavy lifting? Do they just expect it to be served up on a plate and they're looking for the answers and the root? Or, Claire, in your experience, when you are dealing with customers, are they quite defined? Does everybody know what they need to be seeing?

Claire (07:50):

You will get some of them that do, they know what they want. Gary actually mentioned that earlier. Some people it's just uptime, downtown, storage, there's loads of things. But others, they want you to tell them, they want you to leave, they're paying you to manage their service, they want you to tell them what is the best-

Graham (08:06):

So being proactive.

Claire (08:07):

Yes, exactly. They just want you to tell them what's the best thing for them. How are things going? Is everything okay? Sometimes, that's what they want to know, "Is everything okay?" "Yes, it is. And all the graphs are very pretty. It's fine." They want us to make them aware when something isn't right or if we are concerned about a thing. I think, again, Gary mentioned that earlier, if something's not right, we need to keep an eye on it, we need to monitor it and we then need to go to the customer about it or just tell them that that's what we're doing. So you get those that know exactly and those that just want to leave you to it.

Graham (08:37):

I guess.

Claire (08:37):

Someone to take it off their shoulders.

Graham (08:39):

Yeah, I guess that they want to be just in their business and what they do best, which is-

Claire (08:44):

Sometimes, yeah.

Graham (08:45):

... which is not running IT infrastructure.

Claire (08:46):

Exactly. Sometimes they're swamped with information, they don't know what to do or what to look at properly and how to manage it, so that's what they're paying us for.

Gary (08:53):

I think consumer life drives expectation within business life. And I think when it comes to data, we might not be that aware as to how much data has an influence over our kind of day-to-day life because it's done really, really well.


So a good example is with smartphones, I get into my car on a certain day and platform provider of choice says to me, it's going to take you 20 minutes to get to this particular place. And it's done that because it knows that every other time I've got into my car on that day, at that time, I've gone to a specific place. And so, it serves me this data up automatically without me having to ask. And I go, "Oh, that's useful," or "Oh, that's not useful," depending, but it's served to me within context.


So I think that in and of itself can breed a little bit of laziness in that there is an expectation that the right data will be presented at the right time in the right format. And that's where we as providers have to get better. So we need to be providing our clients with the right information at the right time. And as Claire said, when we have these monthly reports, it isn't just about providing, "Here's all of your data," which naturally we can do, if that's your requirement, we can give you all of the data that we've collected and let you do your own analysis. But for us, it's about providing the pertinent data at the right time in the right context and presented in a manner that can be consumed.

Graham (10:29):

And I guess to just follow on to your analogy, because something's just gone off in my brain there about trust. So trusting the data as well, because how many times do we get in our cars and we set the car sat nav, but then, we might, on our smartphones, go to Waze, or we might go to Google Maps? And are we necessarily trusting the data that's on our car and do we trust Google Maps better or do we trust Waze more, for instance? So I guess trust is a big thing when it comes to delivering that data at the same time, yeah?

Dan (10:58):

I think that's trust in the analysis of the data. So the data points are the same, it's starting at this position, you're going to that position, and the traffic along the route is ups, downs resource. We've all seen the greens and reds on those little maps, but it's how much do we trust the analysis of that data to tell us which is the correct route to take?


And I think we mentioned earlier, the whole point of having a managed service provider is a trusted partner within your business. You don't have to spend days and hours looking through logs, looking through graphs, because our customers are trusting us to do that on their behalf and to understand their business context so that we can tell them proactively when we think adjustments to their solution, adjustments to their infrastructure needs to take place.

Graham (11:50):

But we still need people involved, don't we?

Claire (11:52):


Graham (11:52):

Yeah, that's the thing. It's not all been done through AI at all. So what happens when people are away? They leave, they go off sick, where's that leave everybody? Because we still need that humanistic or that human interface, don't we?

Gary (12:06):

That's why, or it's one of the big factors when coming to a managed service provider, you are buying the use of a pool of resource that is far, far bigger than you could fund yourself within your business.


The number of organisations that have a right-sized IT team, okay, so you may have two infrastructure people within your team because one's a single-point failure obviously, so you have two because then that gives you that sort of resilience as it were. What if one of those two people is long-term sick? Or what if one of those two people is on leave and the other person's sick? You've lost that redundancy. So okay, well, then we'll get four infrastructure people and then, we'll get eight. It's a ridiculous example because that's just not financially viable within most businesses.


And that's where we come in is that we have, I think, it's a few more than eight infrastructure people that you can call upon as and when they're needed. So in the same way that we rightsize the compute and the storage, we rightsize or enable you as an organisation to rightsize the amount of resources that you need.


And again, that's part of collecting data from our clients. When are your key events during your year? If you tell us that you are planning on a huge marketing push, we will make sure we have the right resource ready for your business when you need that.

Graham (13:41):

And Claire, from a customer success perspective, do you see over the time that you are involved with organisations, do you see a lot of movement of people are coming in, coming out of the organisation? And does that change the perspective of how they're looking at data, who's responsible for looking for the data, what do they do with that data?

Claire (13:58):

Yeah, you do.

Graham (13:58):

Sort of volatility almost.

Claire (14:00):

You do have that. And that is very hard because there's no consistency if you've got that within the clients. I mean, ourselves, it's slightly different, but when you've got that from a client perspective, that is very, very hard. But we are consistent, so we would be consistent with the information and we have records of the information, but it's hard when you've got it from the client side as well. And that's happening a lot, especially at the moment. There are lots of chopping and changing going on at the moment. So it is very hard. And it's difficult for them as well because that person hasn't got the experience of that company. So you've got the new person coming in, they haven't got the experience, you are almost having to tell them what is happening and how it's been happening and so forth and so on moving forward [inaudible 00:14:39], but yeah, it is a bit difficult.

Graham (14:41):

Well, I think that's a wrap for this month's Fasthosts ProActive Podcast and hope that everyone that's listening, you feel a little bit more up to speed and informed in just all the issues around data and what you should be or what you shouldn't be doing with it.


What's coming up next time I hear you ask. Well, I'm quite excited about this because we're going to be doing a podcast on myths busting. So we're going to be trying to address all those different myths that are out there in the industry right now. And I'm sure everybody here and perhaps a few more people are going to have some input into that.


But for now, Claire, Dan, Gary, you've been great again and thanks for all your time today and until next time, have a great month everyone.

Intro (15:24):

Thank you for listening. We hope you enjoyed this episode. You can subscribe on Spotify or Apple Podcast or visit proactive.fasthosts.co.uk for more info. See you next time.

Orlaith Palmer

Read more posts by this author.