Tag: learn

Omnichannel & Sentiment Analysis (II)

I’ve previously touched upon sentiment analysis within Omnichannel in several articles (https://thecrm.ninja/omnichannel-sentiment-analysis/ and https://thecrm.ninja/omnichannel-supervisor-tools/). It’s really a great feature that allows agents to quickly & easily see how the customer is interacting. It also allows for supervisors to see at a glance how interactions are going overall.

With all of that, I thought it would be helpful to take a further look into how sentiment analysis actually works, so that we can understand it a little better.

Now, the actual nuts & bolts for sentiment analysis are provided by Azure Cognitive Services. There are a wide range of tools available through this, but we have no need to go into Azure to configure this. It’s a simple setting within Omnichannel to get it working, rather than needing to fiddle around with many different things:

However, what’s actually going on during a conversation, and how is the sentiment analysis worked out/calculated? We see the pretty little face icons (with the different colours), but how are these actually being set?

Well, there are two ways in which algorithms are used to calculate the sentiment that’s shown:

  • Natural language processing (NLP)
  • Machine learning (ML) algorithms

With these two ways methods, it’s possible to not only see what the current interactions are showing, but also to enhance the model to understand sentiment better.

Note: In a session that I presented recently, one of the attendees asked if it’s possible to train the model, to result in a custom algorithm. Unfortunately this isn’t possible to do – the machine learning that takes place is the general Azure one, rather than one for a single company or customer

The following diagram shows the sentiments that are used. They’re nicely colour-coded, for ease of reference as well:

When a customer interacts through Omnichannel, the sentiment shown is based on the last 6 messages received from the customer. As a result, the sentiment shown can very well fluctuate & change during the conversation, based on how it’s going.

The Sweetest Languages in the World - | Beyond Exclamation

Obviously, customers aren’t just going to use English to communicate. Companies are based around the world, and will use their native/local language when providing support. Omnichannel allows for this without an issue, utilising the Azure Text Translator API behind the scenes to provide this. If you’re interested to see which languages are supported for this, head to https://docs.microsoft.com/en-us/azure/cognitive-services/translator/language-support which is the latest source of information for this.

There are some interesting things to know around how this actually works:

  • When a language other than English is used, the Text Translator API translates the text to English, and then it’s analysed/scored for sentiment
  • If a language isn’t supported by the Text Translator API, it won’t be scored
  • If profanity (eg a swearword) is detected, the sentiment will automatically be shown as Negative or Very Negative, regardless of the rest of the last 6 lines of conversation

Some people have expressed their concern to me around how accurate the Azure translation actually is, but to date I haven’t seen any major concerns resulting out from it. As with the other Azure services, Microsoft is continually refining & improving it. That being said, there are several languages with very nuanced terms. I’d like to think that these would be supported without issues.

There is, however, somewhat of an interesting behaviour when starting off the analysis at the beginning of the conversation:

  • If the initial language is detected as English, it’s assumed that all of the subsequent conversation will be in English. As a result, if the customer switches away from English, the system won’t recognise this, and a Neutral sentiment score will be shown
  • If the initial conversation is not in English, then the system will check every conversation line & re-detect the language as necessary.

This seems somewhat strange to me, as I’d have thought that the system would automatically check the language for each conversation line. I can think of plenty of scenarios where different languages are used in a single conversation, even if it does start with English being used. I’d like to think that this will be updated at some point, to make the experience better.

Workflows & Managed Solutions

This is about some interesting behaviour around workflows & managed solutions, which I’ve recently discovered. Let me give a bit of background first.

Currently I’m working on several COVID-19 apps for local authorities, to be able to help them assist people in need. As part of this, each local authority has a portal within the solution. The portal itself is a Power App Portal, and I haven’t really had exposure to them before.

blog.atwork.at | Hello, PowerApps Portals (and external users)!
Default portal view, not the one we implemented!

Installing a Power Apps Portal comes with quite a large number of solutions in order to get it to work. More on this below.

Due to the way in which we’re engaging with our clients, the solutions are built in a single tenancy (different environments, of course!). We’re then inviting the users in as guests through Azure Active Directory, to be able to access functionality etc. This works well – we don’t need to worry about managing user accounts, AAD permissions, etc. However it also means that we don’t have any Office 365 licenses within the environment itself.

Now we have workflows that are sending emails out around the portal – registrations, password resets, etc. These are being generated automatically by the system, but as there’s no Office 365 mailbox for the user, they’re queuing up.

It’s not possible to authenticate a mailbox belonging to an external user (we tried!), as the system needs a native (full) user with an active mailbox to be able to send out emails. This is of course unlike Power Automate, where you can create a Send Email action and use specified credentials for logging in to send an email.

So, we did what any normal system administrator/configurator would do. We opened up the relevant (managed) solution, and from there opened up the workflow that we needed to modify. Things looked normal at first – we deactivated the workflow, and started poking around it to see what made it tick.

We came across the part that actually took user credentials to send the email that was being generated, and modified this accordingly. Then we saved the workflow, which was successful. However, upon trying to then reactivate the workflow, we got the following error message (helpful, isn’t it!):

Nicely it gives the option to download the log file around the error. This can usually be quite helpful (at times), so we thought we’d take a look at it. Behold the following (I’ve had to shrink the screenshot to allow it to fit on the screen!):

Isn’t that ‘beautiful’. Don’t worry if you can’t actually make out the error information – none of it makes any sense, at least not in a practical sort of way.

Being stuck at this, I thought to reach out to one of the community Power App Portal champions, Mario Trueba. I’ve known him for a while, and he’s just simply amazing. Having asked if I could jump on a call with him for 15 minutes to diagnose (& hopefully find an answer!), we spent almost an hour!

He suggested trying to use the classic interface, as I had been doing all of this through the new UI. So off I went to open up Classic (I’ve missed this, I will freely admit). Through there, we opened up the solution, opened up the workflow, and re-activated it. Or not, as it happens – even through the Classic UI, we weren’t able to do so. We tried a variety of things, but to no avail. It just simply wasn’t happening!

I was slightly concerned that there was an underlying issues with Portals, perhaps from some legacy CafeX code. I had tried searching with Mario for error details contained within the log file, but we couldn’t find anything that would fix it.

The next morning on waking up & checking Twitter, I noticed someone tweeting around Portals, and engaged with them. They turned out to be on the Portals development team, and told me to shoot them over an email with the details, which I did. They then replied to me, saying that it wasn’t anything specific to Portals, and that I should raise a support ticket. That crossed one item off my list (a Portals issue), but I was still needing to get things resolved.

So I went off & raised a support ticket. A few hours later, a very nice tech support person called Siva gave me a call to discuss the issue. We hopped into Teams, and in what I can only describe as the SHORTEST period of time that I’ve ever experienced, the issue was resolved (it took 7 minutes in total. Yes, I know…). Don’t worry – I’m not going to leave you hanging here!

See, what the ‘issue’ (and I’m deliberately putting it in quotes) was turned out to be something quite simple, yet quite strange.

Essentially opening the workflow from the managed solution somehow (& I don’t know HOW) inherits the ‘managed’ property. This is whether we open it from the new UI, or the classic UI. As a result we’re able to deactivate it, but we CAN’T reactivate it due to the system thinking that we’re modifying a managed component (as an aside, it is interesting how I did manage to save it though?). This was what was causing things to fall over, and the error message was really not helpful at all.

It’s also not a matter of being a Microsoft (or ISV) managed solution. I’ve replicated this happening with a solution that I’ve built, exported as managed, & then imported.

So how did we do it? Well, there are two ways in which this can be dealt with:

Either we can go to System/Processes, find the workflow there, open it up, and then reactivate it:

Or we can open up the Default solution, navigate to processes, select the workflow, and then reactivate it:

Both methods work just fine, and as mentioned earlier on, I’ve since replicated this on workflows in other managed solutions.

To me, this is somewhat strange, and should work regardless. According to Siva, it’s the desired system behaviour, though I have no idea why someone should want it to work in one way, and not in another.

So if you’re reading this, and you might just happen to know someone in the necessary Microsoft engineering/development team who’d be able to answer this, could you point them my way? I’d love to engage them to find out why, how, and if they could pretty please change this?

Omnichannel Supervisor Tools

While I’ve covered a lot of different pieces of functionality from both the end customer & agent experience, I haven’t really touched on the Supervisor experience to date. With the release of a nice little piece of functionality for it around Sentiment Analysis, I’ve therefore decided that it would be a good time to take a look at it.

Now obviously supervisors exist, in the sense that any contact centre would have managers in place to manage the agents. Indeed, they may actually handle customer queries directly as well – miracles have been known to happen! 😉

Really the aim of the supervisor role within Omnichannel is intended to help customer service managers or supervisors enhance their team’s performance and improve customer satisfaction. They need to ensure that customer queries:

  • Are being responded to quickly within an accepted period of time
  • have agents that are providing the right information necessary to handle the query
  • Are handled in a satisfactory way
  • Leave the customer feeling positive & satisfied around the interaction

To this end, the supervisor role includes access to various dashboards & views on the sessions that are taking place. From this, they’re able to ensure that KPI’s are being met, that queues aren’t overflowing, and that agents are carrying out the right work.

Please note that PowerBI Pro or PowerBI Premium licenses are required for users with the Supervisor role in order to access & view these dashboards

On-going Conversation Dashboard

The on-going conversation dashboard is one of the first tools available to supervisors. In this, they can see the entire ‘landscape’ at one glance, covering:.

  • Which conversations are currently active
  • The queue that the conversation has come through
  • The channel that the conversation has come through
  • Which agent is currently handing the interaction
  • The sentiment of the customer experience (more on this below)

Through this, there’s a wealth of knowledge that the supervisor can use. If a conversation doesn’t appear to be going well, they can reach out to the agent who’s handling it, and ask if they need assistance. They can be ready to offer escalation resolution should it be needed.

Supervisors are also able to allocate any conversation that’s an in ‘Open’ state (ie it hasn’t yet started) to a specific agent, or to a specific queue.. They would do this by selecting the conversation in the window, clicking the Assign button (which appears when the conversation is selected), & selecting either Agent or Queue. Depending on which option is selected, they can then select the agent or queue that they wish to.

Furthermore, supervisors are also able to monitor any conversation that is happening (as long as it’s in the Active or Wrap Up stage). When doing this, the customer & the agent don’t know that the supervisor is monitoring the conversation (unless the agent requests it, of course). Through this, they can keep an eye on how things are going, and if they feel it’s necessary, they can join it (without needing to be invited to it). Of course when they join the conversation as an active party, the customer is then notified about it.

To do this, a supervisor would select the conversation (providing it is in the ‘Active’ or Wrap Up’ state) and then can click ‘Monitor’. This action triggers a new session for the supervisor to be pulled into the conversation.

Sentiment Analysis for Supervisors

So one of the really cool features (in my opinion) that’s recently appeared is the ability to use sentiment analysis to automatically alert supervisors.

You’re able to choose the level of sentiment that you want supervisors to be alerted on. When a conversation reaches the predefined level, the supervisor will get a prompt like this:

Sentiment analysis notification

They can then click through to either monitor the conversation (& help out if necessary), or to ignore it. This can be really helpful as it’s allowing the supervisor to be proactive, rather then needing to keep a close eye all the time on the supervisor dashboards. Indeed for a large contact centre, there could be dozens or even hundreds of conversations happening at the same time, so this will immediately flag these conversations to them.

Intraday Insights Dashboard

The Intraday Insights dashboard gives information around KPI’s that are in place for the company. This does need to be configured before supervisors are able to access it – I’d recommend looking at the extensive guide on how to do this at https://docs.microsoft.com/en-us/dynamics365/omnichannel/administrator/configure-intraday-dashboard-supervisor.

There are two sections that are included:

  • Conversations insights
  • Agent insights

Conversation Insights

This section gives supervisors insights into conversations that have occurred over the last 24 hours, against a range of KPI’s. This is therefore a ‘rolling state’ dashboard, without the ability to go back further over time, or pick a specific date-range. If this ability is needed, then the Agent Insights dashboard will be used.

Conversations insights dashboard

As you can see from the image above, we’re looking at metrics across different sorts of KPI’s:

  • Average wait times
  • Average handling times
  • Conversations exceeding specified criteria

This can be very helpful on an on-going basis to see how the contact centre is performing overall, over the time period. Drilling down into these items can help identify bottlenecks & choke points, which can then be worked on to provide better service & efficiencies.

Agent Insights

This part of the dashboard gives information across various health & status items on any given day. Using this, supervisors can drill down into issues that may not be visible through the conversation insights dashboard.

From the screen above, we can see that things like conversation states, agent status & participation modes are available to view.

Because these are PowerBI dashboards, it’s possible to drill down further into them, filter by item (eg queue, agent, etc), and see very specific stats about items.

I hope that this proves to be helpful in understanding some of the tools that are available within Omnichannel for supervisors, and the help that they can give in running an efficient and productive contact centre!

Updating User Settings with Power Automate

Here’s a scenario that could be all too familiar to us. We’re on-boarding users (to either Dynamics 365 or a Power Platform app), & they’re new to the environment that it’s deployed to. So they’re set up, and all ready to go. Suddenly they start asking why records created (or modified) by colleagues show up as having the wrong time on them.

Reverse Wall Clock Unusual Numbers Backwards Modern Decorative ...

Does this sound familiar? I’m sure it does to quite a few people out there!. See, there’s no way to set a default system-wide time zone in Dynamics 365 (or Power Platform). At least not that I’ve come across – if you know of one, please comment below with instructions as to how to do this!

As a result, users are given the default timezone, and need to change it. This is easily done through the Personalization settings area in the app. Users click here, and then select their appropriate time-zone. Brilliant…or so you’d think.

See, when it’s one or two users, it’s generally OK to tell them to do that. However, when it’s 200 or 2000 users, you’re going to get push-back. The last thing you want is for a large number of them to start contacting you to work out how to do it (read the instructions, perhaps?).

User queue stock photo © zam ri (OneO2) (#258450) | Stockfresh

I’ve had this scenario over the last week, where the client actually told us that they didn’t want us to tell users to update it manually. They wanted a better solution.

Well, there is a solution out there to update users. It’s the ‘User Settings Utility’ app that’s in the XrmToolBox (https://www.xrmtoolbox.com/plugins/MsCrmTools.UserSettingsUtility/). Really neat & nifty, and does just what it says on the box. Simple enough to select users (or all of them at a time), select the time-zone you’re wanting to apply to them, and click a button. Hey presto – it’s been updated

Hmm. But what if you didn’t want to have to do this manually. Or (and this is what I was dealing with), there were decent enough number of users being added to the app every few days, & I didn’t want to have to do this as a manual task.

So I started digging into how the time-zone setting was actually stored. It turns out that there’s an entity called ‘User Settings’, which is associated with a User record. Oh, and if you’re going to want to take a look at this entity to see what it contains, it’s NOT available through the front end. You can’t go into the entity list and just display it (though if you’ve found a way to do this through the Power Platform NATIVELY, drop me a line, please?).

Anyhow, back to things. There’s a value for ‘TimeZoneCode’, which maps to a specific time-zone. Aha, I thought! Right – now what’s the best way that I could work out to do this automatically. Checking in with some contacts in the tech community (thanks BlackOps etc!), Power Automate was suggested, so I started to see about how I could go about it…

So, I created a Power Automate Flow (haha…I got the name right there!). On creation of a new user record, it would programmatically go away and update the value to the one for the time-zone that I wanted it to be set as. This actually worked really well.

The only drawback is that through the user interface, it’s not actually shown as being updated, though it has been. Or sometimes it changes, but doesn’t reflect it accurately. This is somewhat annoying, and caused me quite some confusion between checking the front end to see if things were working, & confirming through the back end (& opening records up) to see that it was. I still have NO idea why this was happening.

Before changing my settings
After changing my time zone to USA (EST)

For my specific scenario, all of the users are in the UK, so I set it to update every user on creation to the UK time-zone. Obviously if you have users in different time-zones, you’d want to set this differently. This shouldn’t be an issue though, as you can expand the Power Automate Flow and add logic conditions/branches to be able to do this.

Now I think that this is pretty cool, and I couldn’t find anything out there for this. I’ve therefore decided to release this in a small solution, for others to be able to use. Part of this is the entire list of time-zones with their specific codes, so that you can update to whichever one you need to.

Update Timezone for New UsersDownload

I hope that this helps solve a small but annoying problem (at least it did for me). Please do provide feedback if you want to!

Omnichannel – Pre Survey Responses & Routing

I’d like to start off here by admitting that in a previous blog post that I put up, I mentioned that it’s not possible to route customers to different queues through the chat itself. That was wrong – thankfully several very nice people at Microsoft reached out to let me know how it’s done (thanks BTW for reading my blog!). I therefore thought it would make a good article, as people do ask me about this from time to time.

So, how exactly does Omnichannel facilitate this? Well, there are two parts:

  • Pre chat surveys
  • Routing rule items

Pre chat surveys

These surveys are really quick & easy to set up (or even more complicated, if you so desire). To start getting to grips with them, open a Live Chat record, and go to the ‘Pre-chat survey’ tab

Here, you’ll be able to set up your questions, which is done by clicking the ‘Add Question’ button. When you do this, you’ll get the following prompt.

So, three of the four questions are really quite simple. You need to give it a name (as every system record needs), the actual question text, and whether it’s mandatory or not. The fourth question ask you what sort of question type you’re looking for. The options available are:

  • Single line of text
  • Multiple lines of text
  • Option set
  • User consent

If you select ‘Option set’, you’ll be prompted to enter the values. These should be separated by a semi-colon character:

With our pre-chat survey questions being set up, let’s see how we go ahead and use them for routing.

Workstreams

If you go ahead and open up any workstream record, you’ll see several tabs available. Two of these tabs are Context Variables, and Routing Rule Items. There’s usually one workstream per chat channel, with setting options within it as required. Opening up the workstream for the Live Chat, we can see them there:

Let’s take a closer look at the Context Variables first. Going to this tab shows us the following:

Woah. Where did those entries come from? I didn’t enter anything here – though I can create context variables if I want to.

Well, remember those pre-chat survey questions that we created? Each time one of these is created, it creates a context variable record for the workstream that the chat is associated to. So each of my questions (and I have four of these) now have a corresponding entry.

OK – so the system does that. But how does that help me when looking at trying to route things?

Simply put, these are the building blocks that we’ll set up in the Routing Rule Items to flow the customer chat through to an appropriate location. Let’s go and create one to see what happens.

We need to set the queue that this rule to apply to. Then we’ll go ahead and set the condition/s that we’re wanting to apply for this queue. There are several different possibilities to start with:

Selecting the entity that we want to use for the rule will then allow us to pick an attribute for that entity. So;

  • Account, Contact, Case & Live Chat Context will give an option to select one of the attributes from the entity
  • Context Variables will give the available context variables to choose from

You’ll then be prompted to select an Operator. These will vary depending on the type of field (eg a number field will have additional options such as Greater Than, Smaller Than, etc)

Finally, you’ll enter the value that you’re looking to match with for the condition. This is free text (it’s not auto-populated with values). So in summary, you’ll have something like the following:

And tadaa! it’s active. Brilliant!

We’re able to stack up multiple conditions to cover specific scenarios. An example could the following:

  • Customer has a Kawasaki motorbike (not a different make)
  • Customer’s annual spend falls into the ‘high spend’ bracket

There are plenty of other scenarios that can be covered, and the conditions allow this to cover quite complex situations.

So, some things to note around workstreams & routing rule items:

  • You can have multiple routing rule items per workstream, each one routing to a different queue. These are evaluated in the order that they’re saved in. Eg if there are 4 rules, an incoming chat will be evaluated against rule 1, then rule 2, etc
  • When a routing rule condition is met, the chat gets routed to the destination. No other evaluation against the remainder of the rules is carried out

I hope that this has come in useful, and put some interesting thoughts into your mind as to how you could implement this at your organisation or clients!

Omnichannel Install/Update Errors

I’ve had an interesting time over the last week or so. Several people have contacted me about trying to either install Omnichannel, or upgrading to the latest version. These differ based on what the user was trying to do.

When trying to install into a new environment, the error says ‘To add this channel, you must have an active subscription to Dynamics 365 for Customer Service Chat or Digital Marketing’. This is especially strange as a trial environment (for testing purposes) doesn’t actually require these licenses. It only requires a Customer Service Enterprise license.

When trying to upgrade an existing environment, there’s a different error. This one says ‘We are unable to check for upgrade as you don’t have the required permissions. You need to be either a global administrator or a Dynamics 365 service administrator to check for upgrade. Transaction Id: 0cc1f6be-32f1-476c-8071-acc4d8475e63’. However the user has the Global Administrator role (which obviously also includes the Dynamics 365 Administrator role as well!

image.png

Now I love being able to share my knowledge & help others. That’s one of the main reasons why I started this blog and why I share information that I feel is helpful & useful to the wider community. So I was more than happy to try to help the people who had reached out to me, and jumped on a screenshare session with them (using Microsoft Teams, I may add!).

They were indeed getting the errors mentioned. Nothing that I could suggest helped to rectify. To try to diagnose & compare, I jumped into my own environment. To my absolute surprise, I was greeted by the same issues!

Nanny Knows Best: Shock Horror Probe - People Take Responsibility ...

I knew that it hadn’t been occurring several weeks back, as I had carried out some maintenance work in my own tenant & everything was working fine. I double-checked everything on my end, and it all seemed to be set up correctly.

I therefore decided to go ahead and log a ticket with Microsoft Support. I had a sneaky feeling that it was something, somewhere, to do with the Wave 1 2020 release upgrade. This had happened 2 weeks back (since I had last been into the Omnichannel setup), and I was figuring that something could have gone wrong.

This feeling was boosted by hearing that someone else who was having the same issues had also logged a ticket with Microsoft Support, and they had resolved the issue for the affected tenant. In doing so, they had mentioned that the back-end hadn’t been configured correctly, and got it fixed.

My support agent was a lovely guy called Tomasz, based in Portugal. Emails initially exchanged, we then jumped onto a Teams screenshare session so that I could demonstrate the issues from my side. He was very helpful, and immediately got to work. Within 12 hours I had received an update from him on the situation. They had identified the problem, and were working on a fix.

I had mentioned to him that I knew it wasn’t isolated to my tenant, or even region, but that other people across the globe were also experiencing this. I suggested that whatever fix would be found should be rolled out on a global scale (if applicable).

The crux of the problem seemed to be that with the Wave 1 2020 Release, there had been a change in the architecture of the Omnichannel total solution. Everything still appeared the same through the interface, but under the hood there had been some changes (I have no idea of what actually had changed though).

For new instances (whether Trial or Production), the solution was installing with the new architecture. However all existing systems (whether Trial or Production) had the old architecture, and the Wave 1 2020 Release wasn’t upgrading it to the new one. It simply failed, giving the different error messages.

The fix that was needed was actually quite simple, and only took a few minutes. I had to spin up a new trial of Customer Service within my tenant (which would expire within 30 days). Doing this re-installed the Customer Service solution, & included the new Omnichannel architecture. As a result, after waiting around 5 minutes I was able to open the Omnichannel Administrative Settings, and upgrade my existing Omnichannel deployment. I was also able to deploy to another new environment without any issues. The problem had been solved!

Joyful Green Monster Saying Hurrah Vector Sticker Illustration ...

Overall, this support ticket was an example of how support should be/work. I’ve had times before when it’s unfortunately not gone like this, which makes me value this all the more so.

So, lessons to learn from this. Well, if there’s an issue with deploying Omnichannel to a new environment, or upgrading an existing deployment, fire up a trial of Customer Service, and that should fix it. Brill.

I do wonder how this managed to creep in. Obviously one of the main parts of deploying any new major solution is thorough testing. Perhaps it could be that due to the size of the actual Omnichannel solution, something was overlooked somewhere? It would be good if this sort of situation would be avoided for future releases, and functionality build in to automatically upgrade the Omnichannel solution if it has an old architecture.

Update. I’ve actually had feedback from the Omnichannel team around this. Essentially there’s something different about Trial environments, and this issue only affected them. Production environments (ie with paid-for licenses) wouldn’t have experience the issue. I don’t know why they’re different, but somewhere they are!

Omnichannel – Chat Setup

Looking back at the information that I’ve already posted on around Omnichannel for Microsoft Dynamics 365, I seriously can’t believe that I haven’t already done an article on how to set up a chat channel. I know I’ve talked about some of the functionality within chat itself, but it’s now time to sort this out.

This was the FIRST thing that I did when I got my first Omnichannel environment up & running. The feeling of satisfaction when it was all complete and worked was incredible. I think I may have bounded out of my chair, punching the air!

So, with that all being said, let’s see how to go about it. It’s not that difficult, and there are some helpful settings within it. The functionality has also increased with the Wave 1 2020 release, which is quite cool.

Now, you can create multiple chat channels, and position them where you want to. Each chat channel can point to a different workstream, and then feed into a different queue (more on that in another post).

To create a chat channel, go to the Omnichannel Administration Hub, select ‘Chat’ in the left hand menu, and click ‘New’

You’re then presented with a new Chat record window, to set it up. It’s actually quite simple to go through, with tabs providing different options. Don’t forget about these!

A few things to point out from the main page:

Chat Design

Once you’ve filled in the main information, switch to the Design tab. Here you’ll be able to configure the look & feel of things:

Now at this point in time, you’re only able to use the pre-defined theme colours for the Omnichannel chat widget. That’s not to say that you can’t work around this – if you use an Azure bot, or a custom bot (which needs to be using the Azure bot framework, admittedly), you could set a custom colour there.

You can change the logo displayed – this needs to be a publicly accessible online image. This can result in some fun looks!

You can also set Operating Hours for when the chat will be active (see https://thecrm.ninja/handling-company-hours/ for how to set this up).

Pre chat survey

Heading to the pre-chat survey tab, we can set up survey questions for the customer to answer before the chat actually starts with an agent.

There are some nice options here:

  • Being able to set questions as mandatory or not
  • Different answer types available. Eg text (single or multi-line), option-set, or user consent

Now at this point in time, it’s not possible to use the answers given (eg with using an option-set) to route a customer to a specific queue. It would be amazing if this would happen, but it’s not there yet. Instead the information from the pre-survey questions are displayed in the agent interface. This is aimed at being able to gather information upfront, rather than the agent needing to ask for this during the chat session

Location

The next tab allows the ability to tie the chat widget to a specific website. This means from a security point of view that if someone copies the source code from your webpage, it won’t work on a different website. If no domain is specified, the chat widget can be embedded on any website, without restrictions. It’s a useful concept that can be handy in certain scenarios.

We’re also able to capture the customer geo-location. This will prompt the customer to allow their location to be shared with the agent. If the customer doesn’t consent, then it won’t be shared. Note that this does require Bing Maps to work

Conversation Options

Part of the Wave 1 2020 release has been additional functionality for Omnichannel agents to use. This includes abilities to call, co-browse, and screen-share during customer chat sessions.

I’m going to going into detail around these options in a separate post. I’m also going to be looking into the current solution providers for this, and seeing what each one provides above & beyond Omnichannel integration

Custom Messages

The final tab gives the option to use custom messages for some of the system functionality. Essentially things like starting a chat, ending a chat, and chats timing out all have messages around them.

These are things like ‘An agent will be with you in a moment’:

What custom messages allows you to do is to change these. So for example, you could set up the following to be displayed:

I hope that this has been helpful in seeing how you can set up a chat channel. Stay posted for how to set up the other channels as well!

Omnichannel Desktop Notifications

I’m quite regularly asked various things about Omnichannel. One of the most regular questions goes along the following lines:

Are we able to show a desktop notification to our agents, when they’re not on the Omnichannel Customer Hub screen?

Let me explain what this is all about. When an agent is logged into the Omnichannel Agent Hub, and a new chat comes in, they get the following prompt on their screen:

They can then accept it (which will open up a new chat session), or reject it (which will send it back to the queue).

But if they’re not on the browser tab that Omnichannel Agent Hub is open in, they won’t see any notifications. At all! So they miss out on this, and the customer isn’t engaged with. This obviously is undesirable from a business perspective, as it could even result in losing the customer. So the answer, until now, has unfortunately been ‘No’.

Now in the past when this has come up, I’ve suggested that people take a look at either:

However in my experience these haven’t really been suitable for Omnichannel. This can be due to various reasons, including the client, the requirements, or the infrastructure itself. It’s always been a real annoyance to things, and something that I (& many others) wish would be in place. Several of us have given previous feedback to the product team that this would be really useful to have.

Companies want their agents to be as productive as possible, and this therefore results in a gap in their potential productivity.

Well, the amazing news is that the Product Team for Omnichannel have listened to the feedback given. Not only that, they’ve actually acted on it!

As part of Wave 1 2020 functionality, we now have Omnichannel desktop notifications! This can cover the following scenarios:

  • The Omnichannel agent has minimised the Omnichannel Agent Hub app
  • The Omnichannel agent is working on another tab of the browser
  • The Omnichannel agent is working in another browser window

So what does this actually look like? Well, it’s quite nice & neat to see:

Desktop notification

Very helpfully (in my opinion) it even tells you the browser that’s being used. Users can be running multiple browsers, and this helps as a reminder. If a user has multiple different browsers open, this can assist with working out which one has Omnichannel running in.

Now, there are several different actions that will happen, depending on the agent reaction to the notification:

  • If the agent clicks on the text (but not one of the buttons), it’ll open up the Omnichannel app, and show the agent the notification within the app. They can then choose to accept or reject it within the Omnichannel app
  • If the agent clicks the Accept button, the Omnichannel app will open up & be active, and the session with the customer will start
  • If the agent clicks the Reject button, the notification will go away, and the customer will be returned to the queue

Lets take a look again at the notification within the app itself:

There’s a ‘Wait time’ contained with it. If the wait time expires without the agent doing anything, the conversation is returned to the queue.

This value can be configured by the Omnichannel Administrator, to whichever setting fits the organisation. To do this, go to Notifications in the Omnichannel Administration Hub, open up the notification that you’re wanting to modify, and change the value shown below:

However, you’ll note on the desktop notification that there’s no ‘Wait time’ included on it. This is because the way that notification appears on the desktop doesn’t allow for it to be shown. That isn’t to say that it’s not applicable – the agent will still have the same amount of time to respond. If they don’t respond within this time, the desktop notification will disappear.

Now, there’s still something that the agent will need to do in order to have the desktop notifications to appear. They’ll need to give the browser permission to allow it to happen. The first time that it occurs, it’ll prompt the user as follows:

Allow desktop notification

When the user clicks ‘Accept’, it’ll save the setting, and the desktop notifications will be pushed through. Obviously if they don’t, the desktop notifications won’t appear!

There can be occasions when this still doesn’t work. The below items should help you troubleshoot any these, or similar situations:

One thing that’s also really useful to know is that all of this isn’t just for new customer conversations. The functionality for desktop notifications also covers:

  • Incoming chat conversation
  • Incoming SMS conversation
  • Conversation (work item) assignment
  • Conversation transfer
  • Conversation escalation
  • Conversation escalation from a bot

So really the whole gauntlet of agent interactions that they’d be doing on an on-going basis. Which of course is really helpful, and highly useful.

I’m really quite happy that this has come out as part of the Wave 1 2020 feature items. I’ll be continuing to go into depth around the other functionality that’s part of this release. For the moment, I’m also going to quietly wonder what the product team are going to include next – I’m sure it’ll be very helpful!

Thoughts around the Connection entity

I decided to write this post due to currently looking at the Connections entity. This is for a current project with a very specific purpose. When this came up, my thoughts went back to a previous project some years back when we also looked to use the Connections entity. I therefore thought that it would be good to recap & share my experience.

What are Connections?

Now, the Connections entity truly is a wonderful piece of work. It’s one of the core features that doesn’t actually get much time or effort devoted to it! However, it underpins a lot of the way that Dynamics 365 has been built to work over time.

The best way to summarise Connections is:

Connections are a very easy way to connections records without needing to have to create a custom relationship in the system. Connections can be used between records from the same entity, or from different entities.

See, you are able to connect one record to another record within the system. This could be account to account, account to contact, or contact to a custom entity. There are practically no limits, apart from the extent of your mind! All of this is done by leveraging the functionality that Connections brings to the table.

Note that I’m not talking about lookup fields here, which are also great, but work differently, and require creating a relationship between entities (or even within the same entity).

Just a quick reminder here that custom entities need to be enabled for connections – it doesn’t happen as standard when creating them. You can either do this when creating it, or you can edit the settings for it later:

How to use Connections

In order to connect one record to another, you need to open the first record & click the Connect button on the toolbar:

You’ll then be presented with the New Connection screen, where you’ll select the record that you want. Click the ‘All Records’ item at the top & then ‘Change View’ to select the actual entity that you’re wanting to look for:

You then select the record that you’re wanting, and save. Hey presto, the two records are now connected! To see the connected records, look at the associated ‘Connections’ setting from either record:

OK, so this is really all brilliant. For the absolute majority of situations, it works, and works well. There’s nothing better for it. There are a few small issues, such as the fact that you can’t use Business Process Flows or Business Rules for custom logic, but instead need to use Javascript, but for the most part they work well.

Edge case scenarios & issues

However, there are some edge case scenarios that I’ve come up against, which is the whole purpose for writing this blog post.

What happens if you’re trying to use Connections to establish a hierarchy of records. Eg one record is a parent of another record. Well, you could use a lookup field instead, but if you wanted to define specific attributes for the actual relationship, that wouldn’t work.

Here’s the scenario. You’re needing to capture the relationship between different people, along with certain attributes (eg if they’re a legal guardian, or a trustee, or have power of attorney, etc). You’d think that Connections would work brilliantly for this. After all, you can modify the actual Connections entity to add custom fields onto it. So for example, you could have something like the following:

Note: I’m not referencing Connection Roles, as you can only have a single connection role per connection. In the scenarios I’m handling, I’m needing to have multiple attributes per connection.

So you create the connection between the two records, and you set the attributes that you require. All good. What’s also good to remember is that Connections are bi-directional. You can view them from either ‘side’ of the connection. Eg:

Record 1

Record 2

That’s actually really helpful & useful in the normal scheme of things. You can easily see connections from either side.

But there’s a catch, or even (in our case above), an issue. If we open up each of the two Connection records, we’ll see the following data.

Joe Bloggs connecting to Helen Sommers:

Helen Sommers connecting to Joe Bloggs:

Can you spot the issue? Of course you can! On BOTH of the connection records, the custom fields that we set have the same values. We originally connected Joe Bloggs to Helen Sommers as the Legal Guardian, Power of Attorney & Trustee. Well, if we open up the connection record from Helen Sommers, we’re seeing the same values set, just in the opposite direction!

This is actually due to how Connections work. When you create a connection Record A to Record B, the system automatically creates a mirror Connection record from Record B to Record A. When it does this, it copies all of the values that you’ve set over to this mirror record.

So when you look at the data, you can’t actually see how the structure should work. It’s an issue. Especially if you’re passing the data to other system/s that may need to evaluate it. They just can’t understand this properly, and you’ll get some VERY unwanted results out of this.

Now, there is actually a field within Connections that shows which record is the ‘master’ (ie the one you actually created), and which one is the ‘mirror’ that the system created:

However even with this in place, we’ve found issues when using it:

  • If you’re relying on people looking at the record to see the information, they’re going to make mistakes (ie not checking this value). With the fact that the values are also displayed on the mirror record, this is very prone to user error, and isn’t a good way to do things
  • If passing information to another system (ie the record & the values), you need to program it to only allow it to pass records with this flag set correctly. If the other system is writing back data, it also needs to be configured to write back to the same record.

Summary

With all of this in mind (& especially considering that users may create connections from the ‘wrong direction’, which is quite possible to happen), it’s important to think of the best way to architect systems for regulatory purposes. Financial, legal & other judicatory requirements need to have a system that can handle them properly & accordingly, and not leave room for error.

Therefore, if you’re looking to handle these sorts of scenarios, I’d recommend to look at implementing a custom entity for those specific connections.

Another benefit of this is to separate out these connections from the general connections entity. That way, you’ll also be able to handle security appropriately, which is usually applicable in these sorts of situations. It will allow you to easily allow only a subset of users access (read and/or write) to this data, rather than trying to apply it to Connections (which is going to be a major headache!)

Canvas Apps & renaming field labels

Today I want to share with you something that I’ve realised. Changing field labels can have unintended consequences!

Let’s cast our minds back to the days of ‘traditional’ Microsoft CRM, or as it’s so lovingly referred to nowadays, ‘model-driven’ apps. What you had were a number of entities (eg Accounts, Contacts, etc), all of which contained fields. Fields could be different types (text, integer, boolean etc), and have varying properties on them. You could set them to be required (or not), searchable (or not), and have so much fun.

At the heart of a field is the name that it has. Well, technically there are two names. One is the actual database name. Once a field was created & saved, this was effectively written in stone. The only way to handle a situation where you spelled this incorrectly was to delete it, and then recreate it. Even then, it could still be floating around in the back-end database in its original form.

The second name is the Display name (or Display label). This was the text used on the entity form itself, & could be changed as desired. This was actually really useful – many a time a business unit would say something like ‘we don’t want the field to show as Zip/Postal Code’; we want it to say ‘Postcode’. Well, that was easy enough to address – simply go ahead, load up customisations, & change the display name property for the field. Everyone was impressed & happy, and could get on with their work.

There were of course times that Business Unit A would say ‘I want ‘ABC’ as the display name’, and then Business Unit B would say ‘Ah, but I want ‘XYZ’ as the display name!’. To handle this, it was very possible to customise the label on the form itself, which would then override the display name value. This, of course, would only be valid for that specific form, so it was then imperative to have different forms for the different business units.

Now, in the good old days we use to create SQL queries against the database, SSRS reports, etc. In order to do this, we needed to know the actual underlying (database) field name. We could of course open up customisations, & start trawling through, but there are better methods for doing this. One of these is Level Up by Natraj Yegnaraman. This can be found at https://github.com/rajyraman/Levelup-for-Dynamics-CRM, and is an extension which can be run on Chrome, Edge on Chromium, & FireFox).

Using this amazing tool, it was possible to merely load an entity form up in an existing system, and then TADA! At the click of a button (well, two clicks actually), the underlying database name was revealed. This was an absolute lifesaver, so many times.

So there we’ve been, toddling along for many years like this. It worked, and worked well. All was good.

Then came along canvas apps. Now I’m not a canvas app guru by any means – I’m quite new to them, and still trying to wrap my head around the ‘special’ way in which they operate. Thankfully there are quite a few gurus in the community who have given me help in one way or another to learn how to carry out various functions, and I think that I may JUST be starting to get the hang of it.

With the current COVID-19 situation, I’ve been working on a series of apps for work, to help local authorities. One of these is a canvas app for call centres, to record information easily & quickly. We chose to go down the canvas route due to being able to have a clean layout, as well as being able to display information for the operators to read. This would have been much more difficult in a traditional model-driven app, especially as such things as dialogues have been deprecated.

One of the functions that I’ve had to learn to do this has been to use the ‘Patch’ function (see https://docs.microsoft.com/en-us/powerapps/maker/canvas-apps/functions/function-patch for more information on this. The following is an example of one of the Patch statements that I was using:

This was working remarkably well – it was creating the task record, and setting all of the different values that I needed. For those who are curious as to why I was using a Patch statement, rather than submitting the form, it was due to needing to set the ‘Regarding’ field, which has some very special behaviour!

Then someone on the team said ‘Hold on – we’re only storing one address. Let’s change the field display names to remove the ‘Address 1′ part, so that we don’t confuse users’. OK – I didn’t INITIALLY see any issue with this. I bet that you can see what’s coming through…

Yes – you’re right. The patch statement isn’t referring to the field database name. It’s referring to the field display name! The reason for this is that this is the syntax that Canvas Apps use – there doesn’t seem to be a way to refer to the actual underlying field database name

Of course, I only actually discovered this when I ran through the canvas app again. And indeed, it was whilst demonstrating it to other people! Oh joys – what a wonderful time for it to happen.

So, I then had to figure out what had happened – thankfully that didn’t take too much time. What DID take time was going through every single place in the canvas app that had code referring to the specific fields, and update them to the new (correct) values. This therefore ended up looking like:

So, the vitally important lesson to learn here is be VERY careful when changing field display names, especially if you have one (or more) canvas apps that are referencing them. The last thing that you want is a major headache in having to go back through every place that refers to them, and changing/updating the values.

The only workaround that I’d suggest, is that if you’re wanting to change how fields display in the canvas app itself, change the ‘Text’ value for the field:

That way, HOPEFULLY, nothing will break moving forward.

I hope that you’ve found this useful. If you have a different way in which you’ve handled this situation, feel free to leave a comment below!