Connect Updates Kickoff

Tom Morgan 0:35

It's time for another ACP. We're gonna be talking all things Amazon Connect alex is with me. Hello, Alex. How are you?

Alex Baker 0:42

Hey, Tom. I am good, thank you. How about you?

Tom Morgan 0:45

Good. Yes, very good, thank you. Busy as always, but it's it's always quite good actually to just take a breath sometimes and look back at all the things that are happening, and all the announcements, and all the things that have been going on. And this episode is a good one for doing that, I think, and just remind ourselves that lots is happening. And that's the benefit of a evergreen software, solution like Amazon Connect, is that there's always new things being added. The challenge is then to stay on top of them and work out which ones are relevant to you, and which ones to lean into, and which ones to put on the bookshelf for later, and that's what we're gonna be doing in this episode. We're gonna be going through a couple of changes that have happened in the past couple of months. Feels like maybe these type of catch-up episodes are gonna become actually more regular, 'cause I was looking, 'cause we were talking about what we were gonna do this episode, and actually we've done a, a catch-up fairly recently, but there's still so much to talk about that we didn't talk about last time. You might think a catch-up episode, "Oh yeah, great, let's talk about, the last six months" but actually we're talking about what's been happening in September and August and maybe a bit of October as well. Yeah, there's a lot going on, isn't there?

Alex Baker 1:48

Yeah, when you actually look back over the, all the releases from the last couple of months, it does make you realize, wow, there's, there is, there's always quite a lot going on, isn't there? And you look back and think, "Oh, I remember seeing that one back in August, but haven't really given it much thought." And yes, de-definitely worth talking through some of them.

Tom Morgan 2:05

Yeah. Yeah.

Agentic CX Designer GA

Tom Morgan 2:06

The big one, I guess the, the big elephant in the room is the agentic CX designer. That's been the big kind of hero announcement, if you like, of the last couple of months. It's been in preview for a while. It's now there, it's GA, you can use it. We've got a whole episode about the agentic CX designer, so we're not gonna go over it loads here other than to remind you that it's GA and you can go build on it. If you want to find out more about it, we sat down with Bentley Peyton who is also a cloud interact person, works with us. But he has actually been doing it for real for customers, so he's a really good person to talk to about, the good, the bad, and the ugly, where he sees it going, and where he would use it for workloads today. So have a look for that episode previous episode number 44. So you can go and find out about that. But yes, j- in general, it's out, right? And and I think the line from, from AWS is consider this when building your call flows now. And I think that was Bentley's line as well, right? My big takeaway from that episode was Bentley would consider this the default for making new conversation journeys fr-from here on out.

Alex Baker 3:05

Yeah, definitely. It sounds like that's the way he's he's treating it and where he's focusing his efforts from now on

Contact Lens Data Extraction

Tom Morgan 3:10

Yeah. Other things information extraction for voice and chat. I haven't really played with this, but I think you have a bit, haven't you?

Alex Baker 3:16

Yeah. I've set this up for certainly for a few demos. I don't think we've got anyone using it straight away in production just yet, but I really like this. So w-what it is it's an extension to Contact Lens, and it appears in the contact record underneath the automated, post-contact summary. But you can, via setting up some fields and an additional Contact Lens rule, you can have the analytics capturing key bits of data from the the voice or the chat contact. For example, if you wanted to pull out what was the customer's name, what was the account number, what's the reservation ID all this kind of thing you can fairly easily get that added as sort of individual fields, which is really nice. And then, downstream you could do stuff with it like adding it to a, a CRM field, which again, I've already done in a demo environment a few times. But yeah, really like this one.

Tom Morgan 4:12

This is such an, a quality of life improvement as well for agents not having to type stuff, ask, like actually being able to concentrate on the call, concentrate on what's going on, knowing that some of this key information is being captured for them and not just in a massive transcript they then have to search later. Actually, it's just gonna be, really easy for them to take and do whatever, they need to do afterwards with it. That's really helpful.

Alex Baker 4:35

Yeah, definitely. Maybe some caveats around sometimes the, the transcript can get a bit wonky, can't it? With capturing certain things. And we've had it with particularly one of our healthcare customers where they've got slightly weird medication names. They can come out a bit interesting.

Tom Morgan 4:50

Yeah. Yeah, definitely

Alex Baker 4:52

definitely worth proving it for your own use case. But yeah, in general, really like this one.

Tom Morgan 4:57

And there's an interesting thing about this, I believe, in that this extraction happens before redaction happens. That's either a good thing or a bad thing, depending on your perspective on redaction. But if you need it to drive workflows and drive, whatever happens next after the call, but the struggle is that agents can't then get it out of the transcript because all the sensitive stuff's been scrubbed out of it. This gets you round that because you can set them up as things to be extracted, and they will be extracted before the redaction happens. I guess if you perceive that as a bad thing from a security standpoint, then don't set up the information extraction, and then it won't happen. But I think that unblocked some people who were struggling with redaction because it was impacting the ability for agents to get stuff done,

Alex Baker 5:40

yeah, definitely. We've had it quite a bit with our customers in the past, haven't we, where-- and it's probably a bit of conflicting priorities within the business maybe. So you've got your compliance or security teams wanting to go a bit more full on with the redaction, and then actually maybe your supervisors, your QA teams need to be a bit more pragmatic and, we can't just redact every PII field that's available 'cause we kinda need to go back and figure out what the customer's name is or, certain bits of information. This does seem like, as you say, potentially a nice workaround for that.

Tom Morgan 6:11

Yeah. Yes, that's good.

Agent to Agent Collaboration

Tom Morgan 6:13

I know something else as well, which is a bit techy and involved, but I think is actually gonna be really powerful, which is that Amazon Connect customer now supports agent-to-agent collaboration. So the ability to pass off from one agent to another agent, and it does that using the open A2A protocol which is really interesting 'cause it's got voice support as well as chat support. Chat support makes sense, like lots of agents can do that, but this is for voice support as well. You might have a dedicated agent that does, I don't know like fraud scoring or does loan applications or something like that, and you want to hand off from your generic customer service agent to something a bit more specialist. You can now do that using A2A. And I thought what was really interesting about this as well is that AWS worked with the, the A2A technical steering committee to build in the, the kind of voice type, voice streaming stuff that wasn't in there, wasn't in the A2A protocol, to build in voice streaming so that it can be supported in Connect. So I thought that was really nice.

Alex Baker 7:15

Yeah, definitely. Yeah, it's nice to know that discussions are happening in the background to make it nice and uniform across the, the industry, hopefully

Tom Morgan 7:22

Yeah. And then apparent-- I haven't seen this happen yet, we haven't set this up, but on the contact trace record you've got that l-level of detail about what each agent, AI agent said, and then the tools that got called and how long it took for the handover and everything. So you've got lots of good stuff to help diagnose problems as well. That's, it's all very encouraging actually. Excited to see how that evolves.

Alex Baker 7:43

I guess it does prompt maybe some interesting discussions around who owns the interaction data and stuff, if you're you're involving a bunch of different other external agents. Yeah, so how do you keep visibility of what's going on? That'll be interesting to see how that works in practice.

Tom Morgan 7:59

Visibility, governance, security. Yes all of those things. Absolutely. But yeah I see as well though, as the whole organization becomes AI first, this can really help. So you're not having your contact center go agentically AI in a silo because also your finance team, your HR team, your procurement team, your marketing, what- whatever else it is, they are also getting AI capabilities. And so this helps bridge some of that gap, where you would normally hand off to maybe a human SME for something. Now you can hand off essentially to an AI SME. And then I guess this is That's a

Manager Assistant Insights

Tom Morgan 8:36

good segue. It's almost like I planned it, but I didn't into let's talk about, the other strides that AWS are making in AI for Amazon Connect and for people running contact centers, 'cause I think there's a bunch of new stuff coming for sort of contact center managers as well, right?

Alex Baker 8:51

Yeah, and one that's recently become generally available. I feel like we've been aware of this one for quite a while now because we previewed it with a couple of our customers. But it's the, the manager assistant. You may have seen this appear on the right-hand side when you're in the Connect console. The idea of it is like a, an, an AI chat assistant, which you can just question your contact center data in plain language, get back the answers, the evidence behind it, next steps across all of your metrics. So yeah, that, that's the idea. It's that sort of inbuilt chatbot that you can just, ask, "What was my handle time across these queues for yesterday?" That kind of thing.

Tom Morgan 9:33

I've used this. It's cool. I've only used it on test data, but yeah, I like it. I could see it be- being really useful because it also proactively... it's not a blank screen, which is the challenge sometimes. Everything, like these AI agents, you don't really know what to use them for. They show up in the corner and be like, "You could chat with me," but actually you're not quite sure why you would. Whereas actually it immediately seeds with, you've got three agents online and you've had 50 calls in the last 20 minutes, your average handle time is this. And it's like, "Oh, that's the sort of stuff you can do, is it? Okay, that's cool." That's, and then you can ask it follow-up questions from there.

Alex Baker 10:07

Yeah. It's, it'd be quite a good one to, if you have seen it in your instance but haven't yet had a go with it, just go in and start firing questions at it and try and test the boundaries of what it can do for you. And an interesting example provided by AWS was asking which of my queues are the best candidates for automation. So yeah, that kind of thing where we have our own ways that we've talked to our customers about, in terms of interrogating the data to figure that kind of thing out. But it would be great if you can get some indication of that out of the box and also the justification and evidence that it gives you. I'm quite interested to see what you get there as well.

Tom Morgan 10:45

Yeah, definitely. This has the potential to be, a real change for how people look at dashboards of data and wall boards and stuff like that because you potentially don't really need any of that stuff if if you could just ask it and/or it does a good job of surfacing the right information to you, because it can do a better job of telling you what to do as well as just telling you the data. You might ask it for the data, but actually, really what you're asking for is what is the problem and what should I do about it? And if it can be a bit more proactive about that, then you could imagine just having that on the side and never looking at a dashboard because you've got You know that if there's a problem, it'll come to you and say, hey, this, this queue is getting a bit big. I suggest you do this," be really interesting.

Alex Baker 11:28

That's-- that'll be the, yeah, the next thing for me is the, is it proactive or do you always have to go to it and say, what are the issues in my queues today?" And also can you do, in the, the style of a Claude or a ChatGPT sort of get it to give you a, a daily update or daily roundup of what happened yesterday, what-- where were our i-issues, what do we see happening today based on maybe forecast data how should we adjust our plans? That kind of thing would be quite cool.

Tom Morgan 11:53

Oh yeah. I th-- you know what would be really cool would be that kind of proactive, "Yes, this queue is getting busy. I suggest you maybe move these people from this profile over to this profile because you'll get more people on that queue," blah, blah, blah. "Would you like me to do that? Yes, no. Yes, and also set up a rule to just do it again without asking me next time." Do that for a couple of weeks and suddenly you've got a system that's running itself.

Alex Baker 12:16

Yeah, that would be cool

Tom Morgan 12:17

Really interesting. So yes, that's really good.

Proficiency Routing Reporting

Alex Baker 12:20

The other one that we had in the similar section, but a-around kind of more insight for the people actually running your contact center. So the inclusion of routing steps and agent proficiencies in dashboards. Previously the reporting around the agent proficiency routing was probably a bit lacking and something that you had to expand on with your own custom reporting. But now that the dashboards are giving you the ability to report on those and to, to filter on those. And we've certainly seen a few more customers actually starting to use agent proficiencies. So maybe that sort of uptake in it has driven the need for the slightly better reporting out of the box.

Tom Morgan 12:59

Oh, that's nice. A-again, a really good area where AI could start to handle that. If you do see those bottlenecks because everybody's waiting for a particularly skilled agent that's both a good signal that you need more of those. But also, yeah, does everybody have to go? Could you relax that criteria a bit and cut waiting time? So but even just getting that visibility is really useful.

Alex Baker 13:19

And yeah, it's interesting you mentioned that. That's another thing that some of our customers are keen to explore. Can they be more reactive to things that are happening in the contact center by developing even th- their own tools sometimes to really quickly change proficiencies on the fly and a- adapt to changing demand?

Tom Morgan 13:38

Do you think this uptake of using skills and proficiencies is now that people have got comfortable with how Connect works without it, and they've got comfortable in the routing? 'Cause to embrace Connect properly, you do have to trust in the hands-off approach, don't you? That the routing is gonna work itself a, a little bit, and you can't micromanage it. You've gotta set the rules and the priorities and then trust it's gonna do the right thing which it does. But do you think that maybe people are then adding on skills and proficiencies on top of that now that they've got the confidence and they can essentially make it better again? Or is it just, as these things become more available or just become aware of them?

Alex Baker 14:15

Yeah. So yes, to some extent I think you're right. I think also it's the, increase in granularity of the routing setup that it can give you. I guess arguably the sort of the routing profile where an agent has a single routing profile, unless you go the, the whole hog and have a routing profile per person to make the combinations quite granular. If you don't want to do that and it can be a bit of a, a management nightmare to do that the proficiency-based routing does give you that more, more granular ability to, to tweak, who can handle calls about this thing and to what level can they handle them? So you get that, T- Tom is more highly skilled at sales calls in German. I always want those calls to go to him as a preference

Tom Morgan 15:03

Yeah. Makes sense

Alex Baker 15:04

But yeah I think what you say is pr-pr-probably people are, people have got comfortable with the, the general routing concepts and have found this as maybe a, a solution to wanting that extra granularity

Tom Morgan 15:16

Yep, got it. Thank

Workforce Management Upgrades

Tom Morgan 15:17

you. I know there've been a bunch of changes to workforce management as well. I, I guess we should cover these but we've got an upcoming episode actually with one of the solution architects from AWS responsible for a lot of the workforce management stuff. So we're gonna be talking in lots of detail about some of these things. But we can go over, and I'm just looking down to see if we can touch on the ones that we didn't touch. We've recorded that episode, so we know exactly what's in it. And it's coming soon. But we can touch on some of the things that weren't quite so mentioned in there, can't we?

Alex Baker 15:48

Yeah, it's certainly worth us mentioning them anyway and then, anyone who wants to dive deeper. We've been quite lucky to have yeah we had Pavan a few episodes back talking about FCS, and like Tom mentioned our next guest on one of the upcoming episodes also talks in some depth around FCS. So it's been-- we've been really lucky to get some of the, the real SMEs on the subject to come and talk to us about it.

Tom Morgan 16:11

Yes, definitely.

Alex Baker 16:12

Just a-- so as a quick rundown recently released there was shift bidding, and this is something that definitely comes up in that upcoming episode and we talk about. But yeah, the, the ability for the agents to have a bit more control over th-their own schedule and the shifts that they accept. So that's I think probably something that the Connect community have been calling out for, those that use FCS

Tom Morgan 16:36

Yeah, it always comes up. Yeah

Alex Baker 16:38

Yeah. Yeah, for sure. Then there have been some updates around how you can add unplanned shrinkage, so being able to more effectively model things like when people are logging in late, people that have called in on the morning and called in sick a-and how you can see how that, instantly affects your projected service level. We've also got an update to the refresh of scheduling metrics, which was one in late August. The next one, capacity planning in 15 or 30-minute intervals. Now I'd need to go back and review some of the notes. I kinda thought you could already do that, but maybe it was just for forecasting and now the capacity planning is now 15 or 30. And we have some additional schedule adherence metrics on the dashboard, so more sort of dashboard tweaks bringing more more metrics into there, which is always useful

Tom Morgan 17:31

I know one of the things we talked about on the dedicated FCS episode was actually behind the scenes there's a lot of work that goes on to improve the algorithms as well for forecasting. And so you get the benefit of that if you're using FCS on Amazon Connect as well. And you don't always necessarily see that, and it's not always necessarily talked about but there's a whole team there working to improve the accuracy of those algorithms as well.

Alex Baker 17:54

Yeah. Yeah, that was quite a key point, wasn't it? Very interesting

Tom Morgan 17:58

Yeah. Definitely.

Alex Baker 17:59

Then we have related to FCS and planning your capacity and everything some further changes to concurrency limits. So we had a little while back the, the change that was made so that you could change your concurrency limits per channel and also your after contact work timers per channel as well. You can now set concurrency and interruption limits for specific kinds of task and email work instead of one limit for the whole channel, so you can classify by complexity, priority, or business function. That's really useful for, and we've talked about this in the past where you're using task routing to maybe bring in some of your back office type work and have that all measured and forecasted and scheduled within Connect. If you're doing that, it makes sense that probably your different kinds of work are going to need some, different limits. So it's good to see that granularity. It makes it more of a, a real, prospect to to bring in that back office work into the system, I think.

Tom Morgan 18:58

Yeah, definitely. That definitely feels like a, a listening to customer type feature, doesn't it? 'Cause as soon as you say it, you're like, "Oh yeah, of course." But when you're-- building, I wouldn't have thought of that, as a thing. It's not until you put some of this stuff in the real world and people start using it that you suddenly find these things, right? So yeah, completely makes sense though. It's good to have it

Alex Baker 19:15

Yeah. Yeah, agreed.

Bulk User Admin Improvements

Alex Baker 19:17

And then this one's also really useful a few tweaks to bulk user management. You may have noticed that the UI has changed in a couple of areas actually, so user management, but also the security profiles page has changed the way that you assign the different permissions in a profile. But yes the user management, you can more easily do bulk edits in a more granular fashion. So you can do things like edits to proficiencies, tagging contact handling settings and you can add, remove, partial replace operations. So it gives you a bit more flexibility to do things i-in bulk, whereas it was a bit of an annoying limitation. There were certain things you had to, go in and click into the in-individual profiles and change things one at a time, which is never welcome.

Tom Morgan 20:01

And it's e- exactly the opposite of what you want if you're trying to react quickly to a, like a big surge of calls when actually you just wanna get a whole bunch of people onto a queue or onto a particular skill quickly. Yeah, you don't wanna be doing it individually. That's, yeah, so that's really good. Nice.

Alex Baker 20:17

should also mention around FCS, our friends of the podcast over at Ideal Systems that have that, that great mobile app for Amazon Connect, Mobile Connect it's called. They're soon to be releasing an update for the Mobile Connect app a- and that introduces Mobile Connect FCS, so their own mobile app for Amazon Connect Forecasting, Capacity Planning, and Scheduling, which is another thing that we've heard on n- a number of occasions that people have asked for, even to the point where I recall an RFP that we were looking at where a must-have requirement was a mobile app for the workforce management functionality

Tom Morgan 20:56

completely makes sense. We spoke to those guys back in episode 30, so you can go and find that one. But yes, this is such a no-brainer. So yeah, you can imagine agents wanting to check their schedule when they're at home, on their own m- mobiles. Why not? Same with trading shifts, requesting time off, doing all of those things. Doesn't, make sense for them to do it necessarily in the contact center. They can do it at home. At, and anyway, they might be working from home anyway, but, it just completely makes sense for them to be able to do that, on the move, along with so much stuff gets done on the phone now as well. Yeah yeah,

Alex Baker 21:26

Yeah, that sounds really useful. We'll have to maybe get yeah, R-Reuben on from Mobile Connect, like Tom said in episode 30. Maybe it's time to get them back in to, to

Tom Morgan 21:34

great to get them back. Yeah, absolutely. It'd be brilliant. Yeah, absolutely. Cool. All right we've got more to talk about. There's more things. We're gonna run out of time, of

Quality Evaluations Refresh

Tom Morgan 21:43

course. Quality coaching and evaluation things, so there's a couple of updates there. Evaluations. I love evaluations. I've really got into them in the last couple of weeks and they're really good. So you can now... You don't have to have all your answers add up to 100% now, which is what you used to have to do. Now you can just do it by points and assign points to each each particular answer if you want to. That might help. That just might align better to some of the questions you're doing. And especially if you then wanna get different points for, cutoffs and categorization of evaluations as well. That's pretty good.

Alex Baker 22:14

Yeah. And might just help with the, I suppose the migration piece where you were working in that way previously. You weren't stuck to that sort of rigid percentage scoring where everything needs to add up to 100%. Also in terms of migration, that the next one is quite cool. So you can now upload a PDF of an existing QA form from another tool, and Connect will create it via a bit of AI magic as a draft evaluation form. Not sure if there's any... it does say from another tool, so I don't know if there are any limitations in terms of the format or anything like that. But this sounds like a really cool way of accelerating that, that migration between your QA tools if you're moving to AMS and Connect.

Tom Morgan 22:56

Yeah, it does, and it's much cheaper than writing an integration as well. So that's cool. Yeah, why not? Makes sense, right? Yeah, it's really good. I like that. And I guess, like it doesn't-- it could be an existing QA from another, another tool, or it could just be the spreadsheet that you currently use or whatever, like the photocopied form that you currently use. What-whatever that is, right? Yeah, this can-- you can now give that to, to Connect, and it'll just turn it into an evaluation for you. So that's kinda nice.

Alex Baker 23:22

Yeah, like

Cases and Profiles Enhancements

Alex Baker 23:23

it

Tom Morgan 23:23

Okay, we should talk quickly about cases and profiles.

Alex Baker 23:27

I think it's covered a couple of things. So more additions to the dashboard, which again, always great to add in metrics where they were a bit of a gap previously. So things like case volume, case average resolution time, first contact resolution, SLA achievement, that's from August.

Tom Morgan 23:45

Got it. Got it

Alex Baker 23:46

we can also import via CSV from the workspace for cases now, which yeah, haven't tried that, but sounds like that, that could be really useful. Some of the permissions have changed a-around cases as well. And the ability also to change or add the customer profile after a case is opened, which also is another one of those things where it sounds like probably a bit of a no-brainer if you were comparing cases to another sort of ticketing tool.

Tom Morgan 24:16

Yeah. Yeah, definitely. And again, Cases is newer, than lots of other parts of Amazon Connect Customer. So this is a bit of it just rounding out and getting a bit more mature and filling in some of those gaps as well, I think. So that's good.

Alex Baker 24:28

Yeah. Like you said on the FCS points, it sounds like some real customer feedback there, doesn't it? Somebody said, "Oh, it'd be really handy if it did this."

Tom Morgan 24:35

Yeah. Yeah, definitely. Okay, and we're gonna skim through the last couple because running

Resiliency and Recording Analytics

Tom Morgan 24:40

out of time. But some platform stuff. Global resiliency that sounds quite a big deal actually, or, for kind of contact centers that are really worried about this sort of stuff. Cross-region active-active is now supported across two AWS regions. So how does that work?

Alex Baker 24:56

Yeah, apparently. Apparently and I have to say with our customers anyway, there's been relatively small uptake for global resiliency. My understanding of it is that you had a, an active instance and a standby instance whereas now they can be active/active, and your contacts can route across them. So a contact arriving in one region is offered to the longest available matching agent in either region, which is pretty nice. And it sounds I don't know if it's only specific to global resiliency, but for those, customers that are based across the globe you've always had that slight trade-off in w- if I want a single connect instance, where do I site it to optimize things like latency? Maybe this is a, a good step towards not having to make those compromises and being able to have your agents on an instance local to them, but have the routing a bit more global.

Tom Morgan 25:52

Yeah, that's cool. And both the regions in that setup are taking live traffic. It's not like one's a cold standby, so that's what active/active means. But yeah, it's good. So yeah, that's... I guess the, the people who will find that important will find it very important, and yes, all of the rest of us will be like, "That's great," but it's probably quite expensive as well. Yes. No, that's cool. Yeah. And then some, some, some stuff about external recording analytics as well, so the ability to import voice recordings from other systems and bring them into the kind of contact lens experience, right?

Alex Baker 26:24

Yeah, I love

Tom Morgan 26:25

that's cool. That's nice. Yeah. Yeah

Alex Baker 26:28

'Cause previously, you could do lots of stuff around importing recordings from a, an older or a third-party system. You could bring them into AWS, you could store them somewhere. You could do transcription with some of the other AWS services, AWS Transcribe and and then run analytics on them. What you couldn't do is have them all present within Amazon Connect, and this enables you to do that now. So it's really nice just to have that, one-stop shop for all of your historical data. We've done it in the past for customers. We've imported their data, brought it into AWS, but then you've got the kind of the, the UI problem. How do I, how would I go and search for those calls in a useful way? But this just makes it so much easier. Bring it into Connect, analyze it with Contact Lens. So if you didn't have analytics previously, you can get really good insight into that, that sort of treasure trove of old calls that you've got

Tom Morgan 27:21

Yeah, it's really good. It's easier, it's probably gonna be cheaper, and also means that when you do move over to Connect, you're not starting a completely blank instance, right? You're starting something that's populated and can start giving you insight straight away. Again, coming back to some of that, manager assist stuff as well, it's got all the data in it there to start to help you from day one, so that's really nice.

Alex Baker 27:41

Slight caveat on this, and when we get to the, point of doing this for a customer we'll investigate it. But things like the, the metadata, I'm not sure exactly how you cover that off. I guess it's gonna be limited to what metadata you have from the old system. But then as I understand it, you can import metadata as contact attributes. It's just does that all make it as searchable and make everything as visible as the real Amazon Connect recordings that you have? But w-whatever, it's it's still a great step having them all in Connect, I think.

Tom Morgan 28:13

Yeah, definitely. Definitely. Nice.

Free Sandbox and Wrap Up

Tom Morgan 28:15

Nice. Just before we go we should let you know if any of this sounds interesting and you wanna play about with it but you don't have an Amazon Connect test instance you can try things out on AWS are providing free sandbox environments on AWS Builder Center. I'll put a link to the the blog post about this or the webpage for it in the show notes, but it sounds like a good way to get a sandbox environment if you wanna go and try any of this stuff out and you, but you don't have a, an instance you can do it on today. So that's good as well. Especially with all this agentic AI stuff as well, like really good to set some of that stuff up and just see what's possible and see what you can do, but do it safely in, and why not do it in a free sandbox? So yeah, that's nice. Cool. Very good. We have talked a long time about this. We could talk all day, I think, but it is time to bring this episode to an end. The last half an hour or so have flown by. Thank you very much, Alex. And thank you all for listening. Be sure to subscribe in your favorite podcast player. That way you won't miss the next episode. Whilst you're there, we would love it if you would rate and review us. If you have colleagues that you think would benefit from this content, please let them know. To find out more about how Cloud Interact can help you on your contact center journey, visit cloudinteract.io. We're wrapping this call up now, and we'll connect with you next time