Although CPQ’s Summer ’20 release was delayed due to COVID-19, it’s well worth the wait.…
2. Salesforce Summer 20 Features – Service Cloud, Einstein for Service, Einstein Bots and Field Service Lightning
I hope everyone had a great weekend. Even with the social distancing, it was good to get out of the house a bit as we are finally getting some good weather out here in New England. We actually hit a local state park and hiked in the trails with the family and dog. The dog was 0-3 in the chipmunk chase, but she gave it a good effort. I hope all of you were able to stretch your legs and get out safely a bit as well – it’s been a long time that we’ve been stuck at home. Last week we released part 1 of the Summer 20 release write-up and it focused all on Sales Cloud, Manufacturing Cloud, Einstein for Sales, and the platform updates. Part 2 is always my favorite one – focusing on Service Cloud, Einstein for Service, Digital Engagement, and Field Service Lightning. There’s a lot in Summer 20 here, especially around the new Service Cloud Voice, Einstein, and Salesforce Surveys, so let’s dive in. I have some Broken Bells and Arcade Fire playing so I’m in a good writing mood. As always, the order is based on what I think are the coolest updates and as per lately, the assumption is you’re on Lightning.
Service Cloud and Einstein for Service
- Service Cloud Voice – Let’s start right off with the biggest new feature – Service Cloud Voice. Service Cloud Voice was one of the “wow” moments of the last Dreamforce (I still haven’t quite wrapped my head around there not being a Dreamforce in 2020. I know it’ll be virtual, but still…) and was definitely one of the features that were demoed the most around Service Cloud. With Summer 20 it’s finally here and there’s a lot to it. Just to get it out the way – this is an additional fee – pricing hasn’t been announced yet, but behind the scenes, Service Cloud Voice is leveraging Amazon Connect which charges all usage-based fees. Amazon Connect is a full communications platform with a wide range of services to tap, and Service Cloud Voice is packaging a few of those together. Without a doubt, you’ll see this extend in future releases, and you can tap into those other services if needed now through APIs. For now, Service Cloud Voice starts with agents being able to leverage voice as a channel within Salesforce. Instead of having to connect a 3rd party phone vendor to Salesforce leveraging the Open CTI connector, you can now leverage the Omni-Channel widget to take advantage of voice directly. The Service Cloud Voice connector has all of the features you’d expect from a voice platform – the ability to change an agent’s status, the ability to receive and end calls, ways to transfer a call, and also merge calls. Caller ID is also displayed within the widget so the agent can see who is calling. Behind the scenes, you can set up routing rules and even set up an IVR leveraging Contact Flows and Lambda. These will ensure the calls are routed to the correct, available agents. You also have the option to activate call recording and call transcription services (again, leveraging Amazon Connect services to perform and even store these). Even though the recordings are stored in Amazon, they’ll be linked to the record leveraged for the call – like the Case – so they can be accessed later. Real-time Voice Transcription was definitely the most exciting feature of this that was showed off at Dreamforce, and it allows agents to see a live stream of the call transcript. That transcript identifies the agent and the customer on the call – which is key, as it’s not just a text stream – it’s actually a conversation stream. The big benefit here is that your agent can see what is being transcribed as they go and they won’t need to spend minutes after each call repeating the phone conversation as part of their Case wrap-up. If all of the context is in the transcript – and the agent can see that it is – they can focus on the important unspoken details and potentially save minutes on each call. No more novels written in your description fields. The transcript itself is obviously great for reviewing the case later, or for management performing QA (or even potentially a lifesaver for disputes with a customer about what they said or didn’t say) but the real-time aspect really helps the agent be more productive. Supporting all of this, there’s already a Service Cloud Voice Analytics app that comes with this, which allows supervisors to see real-time data around the call volumes coming in. This leverages Einstein Analytics and can also be customized and extended. This alone is actually a very exciting feature as typically you have voice reporting within your phone vendor’s system, and your Case reporting in Salesforce. With this, you have a combined dataset all within Einstein Analytics. A ton of reporting possibilities there. This is a big new direction for Salesforce, who previously had provided toolkits for the Open CTI Connector and told vendors to have at it. Now Salesforce has its own option. In future releases, they will be opening this new Service Cloud Voice Connector to any 3rd party that wants to take advantage of it, so if you already have a phone vendor they may be connecting to some of this functionality in the future. One big thing to consider when leveraging Service Cloud Voice is the support model. Amazon is making all of these tools and services available for you to use like Lego blocks, but they don’t really support you on it. Salesforce hasn’t announced the support plans yet, but my guess is they have something in mind to ensure you’re covered when moving teams onto this. Overall, this is an exciting new direction for Salesforce and one we will be specializing in as well.As part of our Service Cloud Lunch & Learn webinar series, we will be hosting a webinar focused on Salesforce Voice on Wednesday, August 26th 1:00EST / 10:00PST. I know that seems like a long way out, but it’s only a month after Summer 20 goes live. We’ll be showing a live demonstration on how Service Cloud Voice works and all of the options you can set up. To sign-up, please register here: https://info.gearsautomation.com/salesforce-voice-webinar
- Salesforce Survey Enhancements – A huge release for Salesforce Surveys and some of these new features are big ones. The biggest is a new feature called Customer Lifecycle Maps, which really brings Salesforce Surveys beyond just a transactional survey tool. This new feature allows you to build a map of different points of a customer’s lifecycle with you and to send different surveys at each point of that lifecycle. So, instead of just sending a survey after each Case (which you can still do), you can gather feedback at key points like onboarding, renewal, first anniversary, etc. Included with this is a couple of great new Einstein Analytics dashboards. The Customer Lifecycle Map dashboard (which is down below) gives you great insight into the different points in the lifecycle map and how your customers are responding at each point. It also analyzes your Net Promoter Score, survey engagement rate, and who is participating. Coupled with this is the Response Analysis dashboard that lets you drill down into each question and their responses. The cool part here is if you have the same question in multiple surveys, the dashboard will analyze that question across each survey (perfect for the lifecycle mapping to see if the feelings around a particular subject are improving or declining as your lifecycle advances with your customer). The release notes do say you need Salesforce Surveys Advanced Features permission to leverage all of this – which I think is part of Salesforce Surveys and not extra, but I will try to confirm** (see note below). That would normally be enough for one release, but we get even more around Salesforce Surveys. First, you can now create records based on the survey responses. You can build a quick data map of the questions to field within a record and create records as responses come in. A good example is if you get a bad survey score, you can create a Case record to a supervisor follow-up. On that new Case record, you can populate Case fields with some of the answers from the survey. Next, you can now personalize your surveys with merge fields. This can be done on the survey welcome page/text, or within the questions themselves. Definitely a nice touch with customers, where you can show that you know a bit about them to make the survey a little more personalized. Another new enhancement is the ability to add post-chat surveys embedded within the chat experience. A survey tidbit here if you’re not already doing it – surveys within the chat window typically have a much higher response rate than surveys in any other channel. Essentially you have a bit of a captive, real-time audience as it appears as soon as the chat ends. Just don’t make it a big survey – at best you have two questions – and you’ll get a terrific response rate.[Editor’s Note: Soon after the Summer release, Salesforce wrapped the features above into an add-on licenses Customer Lifecycle Analytics and Feedback Management. Designed as a new suite of enterprise-level tools, please note that there are extra costs associated with these features.]
We also get a couple of new question types. The Like or Dislike question type is exactly what it sounds like – it’s a binary option that lets a customer click on one of the options. This can appear as Like / Dislike, Happy / Unhappy, or Yes/No (perfect for your rapid-fire post-chat survey!). There’s also a new Short Text question type which can collect up to 200 characters. This is great to actually gather information about the participant vs. using for true survey questions. Finally, you get the ability to send out QR Codes with the survey invitation. Participants can now just scan the QR Code and the survey will pop up. One thing to note is these types of surveys won’t be tied to a participant’s record in Salesforce as anyone could scan this. Believe it or not, there are a bunch of other smaller enhancements as well, but I’ll stop here. Huge update for Salesforce Surveys. As a reminder, with Service Cloud you get 300 Survey responses for free, but beyond that you need to purchase more. Surveys are critical for Customer Service, so most likely you’re purchasing a platform somewhere for this. I think with the last few releases and especially Summer 20, Salesforce Surveys makes a very compelling case to be that platform. As part of our Service Cloud Lunch & Learn webinar series, we will be hosting a webinar focused on Salesforce Surveys on Wednesday, September 23rd 1:00EST / 10:00PST. We’ll be showing a live demonstration of Salesforce Surveys and all of this cool new functionality. To sign-up, please register here: https://info.gearsautomation.com/salesforce-surveys-webinar
- Einstein Bots New Features – Einstein Bots also gets a ton of new features with Summer 20. Event Logs are now way more powerful and an invaluable tool to build better bots. The enhanced event logs will now allow you to troubleshoot very complex bots all on one page by letting you see all of the messages sent to and from the customer. These logs will detail the time, the dialog name, the event type, and then the actual event result, which includes the message. Any bot errors that occur will also be logged. With this level of detail, you’re now also able to build KPIs to report against like average conversation time and how customers interact with the bot. Super useful. The Natural Language Processing (NLP) models also got “smarter” with a deep learning upgrade. You just need to retrain your bot with a single click and it’ll leverage these new models. Another big new feature is multi-language bots. Previously, Einstein Bots were English only. Now we get bots that can support Spanish, French, German, Italian, Portuguese in full GA and Chinese Simplified, Chinese Traditional, and Japanese in beta. Huge update for multi-national companies and even for our Canadian customers who need everything in English & French. As part of this, you need to set the language for the bot when you create it, and it’ll be able to handle NLP and menu-driven tasks in those languages. When needing to route from a bot to an agent, Einstein Bots now supports Skill-Based routing so if you’re using that in Chat, you can also have your Einstein Bots leverage it. Yet another enhancement is that Einstein Bots now is fully GA for Facebook Messenger. Facebook Messenger is obviously a huge messaging platform (40 million active business a month!) and now you can tap that with bots as a self-service channel. Finally, the bot builder itself received a few improvements as well. You can now use bulk actions to train your bots quicker and the UI itself has been enhanced to make it faster to navigate. Big release for Einstein bots. Just a reminder, you get 25 free bot conversations a month for each Chat or Digital Engagement license you have. You need to pay for volume above that, but depending on your users, that’s a lot of free conversations to take advantage of each month. It’s part of your subscription, so if you’re a chat user and you’re not using bots, you really should be looking at ways to leverage this to improve customer satisfaction and self-service. As part of our Service Cloud Lunch & Learn webinar series, we will be hosting a webinar focused entirely on Einstein Bots on Wednesday, July 15th 1:00EST / 10:00PST. We’ll be showing off some of these new features and how bots can work in general. To sign-up, please register here: https://info.gearsautomation.com/bots-and-digital-engagement-webinar.
- Chat & Messaging Enhancements – Speaking of chat and digital engagement, they weren’t left alone in Summer 20 and have a few enhancements as well. On the Chat front, skill-based routing just became a lot easier to set up as you no longer need code to map fields from your pre-chat form to the routing rules. You can now set up that mapping using basic setup and assign chats to the correct skill entirely from your pre-chat form. Very nice to see. With SMS Messaging, Canada gets a very big improvement in the way of Short Code support. Canadian SMS laws are stricter than the US and any kind of significant SMS volume requires the use of short-codes instead of long numbers. Previously Marketing Cloud supported these Canadian short-codes but not Messaging. That is now a thing of the past and Messaging can now leverage Canadian short-codes, too. Hourra! (that’s Canadian French for hurray). We also get expanded Consent capabilities, which now support multiple languages. When using different messaging channels or things like short-codes, you will need to support different consent regulations, like explicit opt-ins or opt-outs with universal keywords. These rules allow you to build that out – now in multiple languages – and remain compliant. If you’re using any kind of volume with messaging, this is definitely something to make sure you’re doing. On the customer-facing side, the chat window header gets a new slimmer look and feel. This will happen automatically for anyone using the Embedded Chat (versus the older chat buttons). Finally, the Embedded Chat has some optional animations that can be leveraged including a fade-in and fade-out when you’re minimizing or maximizing your windows. Also, don’t forget about the new post-chat surveys we mentioned above. Last but not least is one of my favorite new features for the whole release – Einstein Reply Recommendations. This came out as a pilot in Spring 20, but it’s now GA. Reply Recommendations takes quick text to the next level by presenting real-time recommended responses to chatters based on the text in the chat transcript. It’s almost like a combination of quick-text and Next Best Action, where the agent is presented with options and can select which text to post to the customer. They can also quickly edit the suggested text – making it more personalized or topical to the conversation – and they can also flag ones that aren’t helpful. The recommendations prediction model will learn from the ones being selected as well as flagged as not helpful and continue to improve their results. Now when this came out in pilot, it was recommended that you have at least 10,000 chat transcripts in your instance for it to learn from. I don’t see that as part of the GA notes, so it’s possible that threshold is no longer needed. Overall though, this is an awesome feature and without a doubt something you want to turn on for your chat users to help them respond quickly and with the proper responses. As part of our Service Cloud Lunch & Learn webinar series, we will be hosting a webinar focused entirely on Chat on Wednesday, June 24th 1:00EST / 10:00PST. We’ll be talking about how Chat and Messaging work and then going into a live demonstration of both. To sign-up, please register here: https://info.gearsautomation.com/digital-engagement-webinar.
- Prioritize Additional Skill in Skill Based Routing – We mentioned that Einstein Bots now support skill-based routing above, but we also get the ability to prioritize additional skills. The way this works is the routing will try to find an agent with all of the needed skills at first, but if no one is available, after a skills timeout period (which you set) it will begin to drop skills that are needed based on your prioritization. This will allow an agent to potentially be found that has the most important skills even if they don’t have all of the skills. Pretty cool enhancement that will help make sure your hold times aren’t too long and work isn’t bottlenecked from getting assigned.
- Determine Agent Capacity based on Work Item Status – An interesting new option for Omni-Channel here. Today, an agent’s capacity is calculated based on what tabs an agent has open. Once that tab is closed, that record is no longer considered part of their capacity. While that does work in most instances, especially high-volume / one-and-done type work, what if your agent needs to go research that work to actually complete it? Maybe they are assigned work types that require them to spend hours outside of Salesforce to solve the issue or research the problem? In these instances, closing the tab doesn’t mean the agent is done – they might just be getting started. Now for these types of situations, you can leverage the Work Item Status to set capacity and not use the tabs. So, if an agent has 5 Cases in a “working” status even though none of them are currently open in the browser, they’ll have 5 cases calculated as the current workload they have. This is an optional setting and doesn’t make sense in a lot of situations, but for those where it does, this really opens up Omni-Channel work distribution to a new type of workload. Love it.
- Einstein Case Classification Improvements – Einstein Case Classification now also supports recommendations for lookup fields as part of the prediction model. This is great for automatically linking Cases to related records like Assets or Service Contracts. Previously all it supports was picklist and checkbox fields. In addition, you now have the ability to set a minimum confidence level for the prediction and when it will be used to automatically update the Case. For example, if the model has low confidence in the prediction, you can set it so those won’t attempt to auto-populate. Previously it was an all or nothing. As part of this you get two more fields that allow you to build out logic and reporting around this – IsUpdated which shows whether the Case was automatically updated or not and UpdatedField which is a list of fields that were updated if it was. Finally, there is a cool pilot that builds off of Case Classification, but this is for when you’re closing the case. It’s called Einstein Case Wrap-Up and it’s for Chat cases. Based on the chat data, it’ll make recommendations on what to populate the Case with when the agent is closing it. Pretty cool, but for Summer 20 this is a closed pilot, so be careful using this in production if you ask to turn it on. As a reminder, you get 1 free Case Classification model as part of Service Cloud, so this is something you can set up and see how your agents use it without any additional investment. This is something a high volume team could really benefit from.
- Knowledge Category Reporting – We wrote about the Reports & Dashboard changes in Part 1 of our Summer 20 recap, but the Knowledge Category reporting enhancement was worth repeating here. With Summer 20 you can now leverage Categories as a filter within reports. Not only can you filter them by the Category, but you get some new operators as well – at, above, below, and above or below. This lets you filter by all categories above, below, or both of the categories you’re adding in as a filter. These filters are available in any report type that has Knowledge Articles. This is a huge improvement for Knowledge reporting and I love the extra twist that Salesforce added here.
- Knowledge Sharing – For as long as I can remember (it was 2008 when Salesforce acquired InStranet, which became Knowledge a year or so after, and I can’t remember this ever being different but I could be wrong) you controlled access to Articles by leveraging Knowledge categories. First, this was always a bit of a pain to do, but second, if you were an author then you had access to all Articles and this didn’t apply to you. With Summer 20, Knowledge Articles can now be shared just like every other object in Salesforce – with the normal sharing features like org-wide defaults, role hierarchies, and criteria-based rules. This deserves a hooray for sure. When you leverage this, the data categories no longer control record access, so it’s an either/or – which makes sense as it’d be a nightmare trying to layer both on at the same time. As part of this, we get the ability to set and change the Owner of the article. One thing to note, this is a beta feature – so definitely test this like crazy before releasing it out there.
- File Redacting of Sensitive Data – We’ve mentioned our GearsDataMask app previously, which protects your Service Cloud from customers sending in sensitive data – like credit card numbers or social security numbers – through the various support channels. When it detects the pattern, it automatically obfuscates the data and ensures it’s not sitting inside an exposed, unencrypted field in your Case data. We now have a way to do this with attachments as well. Sometimes that sensitive information is in an attachment – like a scanned statement or health record – and that is much harder to detect (and easier to miss and have sitting exposed in Salesforce). With our File Redactor we can identify these patterns within Images (think scanned files or faxes), PDFs, Text files or Microsoft documents. Once detected we can notify, delete, or redact that content. All of this can be supported with workflow to build out processes around these attachments. For example, if you need that social security number to process the work the Case is for, we can store it until the Case reaches a certain status and then delete or redact the attachment. Lots of amazing stuff you can do here. To learn more, check out our File Redactor options. This can be purchased separately or used to augment GearsDataMask.
- Embedded Services Channel Menu Enhancements – A cool new feature for the Channel Menu allows you to make it dynamic leveraging a Reorder API. This is actually something we have done custom in the past and is something I really like as a concept. Instead of having a menu of different channels to reach you, that is the same no matter where you are on the website, you can help guide your customers to the preferred channel by re-ordering the channels or even hiding channels based on the content of the page. I’m not describing a way to try and aggravate your customers by hiding your support phone number but rather based on the page the customer is on when they reach out, certain channels may be better than others. For example, if you’re on a page that is an extremely complex Knowledge article or support process, mostly like chat isn’t the best medium. However, if that page is about resetting your password, why even present a call option? Other examples are some pages that might have more sales-type content, so have an option to create a lead. This is a bit of effort to build this, but you’ll increase customer satisfaction by reducing the need to switch channels within an interaction, and you can drive self-service channels in areas that make sense instead of frustrating a customer by forcing it in front of all situations – even ones that are unlikely to be solved with self-service. Finally, similar to chat, we get a few optional animations to enhance how the channel menu works. These animations include a close poof, an action-button poof, and a menu-item hover. Tough to tell what they look like from the release notes, but from a UI stand-point animations do help a user see that their navigation is processing.
Field Service & Lightning Scheduler
- Lightning Scheduler Resource Selection – Lightning Scheduler is built leveraging the powerful booking engine of Field Service Lightning and exposes those capabilities to your customers via the community or a website. When you’re a field service organization, you do want to have the ability to pick and choose the resources to go to a service appointment – but that doesn’t mean you want to give your customers that ability. Two new enhancements give you the flexibility to let your customers just pick a time, and don’t worry about the “who” while Lightning Scheduler will figure that out behind the scenes. The first option is fully anonymous scheduling. This will turn off the ability to select service resources within the flow and your customers will never be able to see the service resource names. A second enhancement gives the customers the option. They will be prompted to automatically assign a service resource or to select one. If they select the automatic option, they’ll set their preferred time and Lightning Scheduler will figure out the specific resource for them. If they pick the select option, it’ll be the normal flow where they get to see specific service resources and their availability. I like the flexibility here. Speaking of Lightning Scheduler, with covid-19, scheduling visits is taking on a much more critical role for a lot of customers. With a lot of businesses restricting onsite access – either entirely or constraining capacity – scheduling visits in advance is more important than ever. Places like banks and retail stores where the model used to be “just show up and someone will help you” now need to maximize their allowed capacity and a big way to do this is by scheduling visits in advance. Lightning Scheduler can be a quick help with this. In addition, we’re seeing companies also trying to schedule virtual visits instead of physically showing up. Using camera share you can schedule virtual shopping sessions or even perform virtual service appointments. We detail out how to build out a scheduled virtual service appointment solution in our latest blog post, but we also offer Lightning Scheduler Quick Starts for in-person or virtual scheduling. If you’d like to talk about how Lightning Scheduler can help your business maximize their time, contact us, and a Solution Architect will get back to you right away.
- Dispatcher Console Enhancements – The Dispatcher Console gets a nice upgrade with Summer 20 that will make it quicker to use. First, you get the ability to select multiple appointments at once and then open that selected list with a single click. Pretty cool. We also get a nice hover capability that lets you hover over an appointment record in the list and see quick information without having to open it. You also can now add icons to the list that can represent different key pieces of information to make it really clear what type of service appointment it is – or what priority it is. Finally, you now have the flexibility to control what users can see and do on the dispatcher console Gantt. You can create read-only versions, or even control which type of users can drag appointments in. One thing to be really careful of here is when you activate these settings it looks like they are not reversible from a global stand-point (you can still change users). For example the notes flat out read: “If you activate these optional permissions without adding them to users, you make the Gantt permanently read-only (Harry note: when a release note uses italics, you know it’s serious). Activate extended permissions only after you add them to users and the FSL permission sets.” Yikes. So, definitely – without a doubt – do this in your sandbox first, read the directions and be careful.
- Optimization Insights – I love this one. The optimizer has always been a cool feature of Field Service, but it was always a bit tough to see what your policy tweaks were doing globally. A lot of it was gut feel. Now with these insights, you’ll get a snapshot as you make changes to your policies and be able to see before and after results of key KPIs like average travel time, total scheduled appointments, etc. It’ll also measure the before and after against your key objectives. These metrics are terrific to be able to track the impacts of your changes as it’s possible you’re actually making things worse without even realizing it. With this in place, you have way more visibility and I’m sure this will keep expanding with more KPIs – making it even more powerful.
- Expense Tracking – Technicians can now keep track of their expenses directly from the FSL app. These are added directly to the Work Order and can be used to track expenses such as travel costs and out of pocket expenses, like if they need to swing into Home Depot to pick up a needed replacement part. For this release, the expense lines are only tied to the Work Order, but you could easily clone these out to create a true expense report that covers a technician’s expenses across all Work Orders for a time period and then reimburse from there. Really nice new feature.
- Asset Availability Tracking – This is a pretty cool new feature. We now get Asset Downtime Period records that can be created and associated to an Asset record. So, if you’re taking down an Asset for maintenance, this can be tracked with a record for how long it’ll be down and when. You can even track if that was planned or unplanned downtime. These records then roll-up to the Asset itself to calculate it’s overall uptime availability. Really cool. It’s part of Field Service Lightning, but to be honest, I could see this being really useful in pure Service Cloud as well (it might be available there, we’ll need to check).
- Criteria Based Sharing for Time Sheets – Time Sheets have been getting more and more features the past few releases and with Summer 20 we get the ability to add criteria based sharing rules to them. Time sheets can definitely be a sensitive set up data that only certain teams should see (we don’t need everyone seeing that John has been slacking this month…) so this is a needed feature to ensure only the necessary users can see this time entry data.
- Manage Loads for Trucks and Workers – We can now add Count work rules to our scheduling policies that sum up the total quantities being assign to a Worker or Truck over a period of time. Sometimes you might be able to squeeze in more appointments from a time stand-point, but the truck simply can’t handle the load for that many appointments. Now you can track this – ensuring that a truck or worker is only assigned to the amount they support in a given trip.
- More Confetti – Confetti is now tossed when your path hits a designated stage on product requests or product request line items now. Sigh
Wow, this was a ton. Huge release for the service side of Salesforce. As we mentioned in a few places, we have a Service Cloud Lunch & Learn webinar series going that highlights specific parts of Service Cloud, and all our sessions include a live demo. Check out the full list here: http://www.gearscrm.com/webinars/ and hope to see you there. As always, if you have any questions or want some help in getting any of these features rolled out in your own instance, please reach out and we’ll get a Solution Architect in touch. Next up will be all of the Collaboration features of Summer 20: Communities, Quip, and Chatter.