Agents and supervisors, the channel add-ons, seasonal headcount, and deflection as a licensing strategy. Three knowledge checks along the way, and 4 clips from a senior licensing analyst.
This is a taught session, not a talking head. The instructor works through analyst grade slides, and three times the video stops on a question with four options on screen. Pause, commit to an answer, and the next slide explains which option is right and why each of the others is wrong. 4 times in the session the frame splits and a senior licensing analyst gives the view from inside real SAP negotiations, and the instructor picks the clip apart when the slides return.
The full narration of this session, section by section, for reading and reference. Guest analyst clips are marked.
Welcome back. Session seven, Service Cloud, and this is an estate with a shape that Sales Cloud does not have. Contact centres have peaks, they have channels that arrive one at a time, and they have a population that looks uniform from outside and is not. Today: the two cloud question and the test that settles it, what the core licence actually covers, how channels price, the seasonal problem that costs more than any negotiation will save, and an argument I think licensing people should make more often, which is that case deflection is a licensing strategy. Three knowledge checks. Let's begin.
Five objectives. First, settle the two cloud question, so when somebody needs both, with a test that answers every edge case your administrators bring you. Second, say what the core licence covers: cases, console, queues, routing and knowledge, and where the boundary sits. Third, price a new channel properly, meaning per agent or consumption, and what happens to the number if the channel succeeds. Fourth, plan for the shape of the year, because seasonal peaks against annual commitments is the single largest recurring waste in a contact centre. And fifth, use deflection as a licence lever, since case volume drives agent count and agent count is the bill.
So, two clouds and one user. Four cases. Somebody who reads a deal, meaning a service agent looking at opportunity history for context, is usually fine on Service Cloud alone. Somebody who works a deal, owning opportunities and running forecasts and sitting in the sales process, needs the sales side. Somebody who takes a query, a salesperson answering the occasional question, does not automatically need Service Cloud. And somebody who works a queue, handling cases to a service level, is doing agent work and is licensed as such. Let me put the test in a form you can hand to an administrator.
Guest analyst clip.
Every single edge case your administrators bring you. And it is worth writing down rather than holding in your head, because the alternative is that each case gets decided by whoever is asked on the day, and the safe answer for them is always to issue both.
Right, what the core licence actually gives you before any channel. Cases and the console: the record, the queue, the agent workspace, the thing agents live in all day. Routing and queues, getting the right work to the right person, which is most of what a contact centre actually is. Knowledge, so articles agents use and customers may see depending on how you publish them. Entitlements and milestones, which is service level tracking and matters a great deal if you sell support as a product. And supervisor visibility over queues, workload and performance, which is usually the same licence rather than a separate one, though it is worth confirming that on your own paper rather than taking my word for it.
First knowledge check. A service agent occasionally opens an opportunity to see what a customer bought. What do they need? A, Service Cloud plus Sales Cloud, since they touch both. B, Service Cloud alone, since reading a deal for context is not working the pipeline. C, Sales Cloud alone, since opportunities are the sales object. D, a Platform licence, since they mostly read. Pause here and pick an answer before you continue.
B. Service Cloud carries the core objects, so context is included, and what it does not carry is the pipeline work of owning opportunities and forecasting. A double licenses somebody on the basis of a screen they look at occasionally, which is how service estates quietly double in cost. C removes the case handling that is their actual job. And D fails because an agent working cases needs the service functionality, which is exactly the boundary session four drew around Platform licences.
Now the channels, and there is a clear pattern. Email and web forms are included with the core licence, so that is your baseline. Messaging and chat price per agent on top of the agent licence, and the question is which agents and whether everybody needs it. Social and WhatsApp usually sit inside the same digital bundle, where the question is whether the bundle costs less than the channels you actually want. Voice is often consumption, priced on minutes, so the question is what the forecast is and who watches it. And field service is a separate product with its own user types, where the real question is whether those people are agents at all or a different population entirely. The pattern to notice: the core licence is a platform, and each channel is priced on top of every agent who touches it.
Which brings me to how this creeps. Nobody frames adding a channel as a licensing decision, because meeting customers where they are is a good reason and not a costed one. The per agent uplift compounds, since a channel bundle across two hundred agents is a percentage on your largest line every year. Voice behaves like Data Cloud, because minutes are consumption, so the bill follows customer behaviour rather than headcount. So you ask before go live, and then you watch it monthly. Let me explain why the shape of this matters.
Guest analyst clip.
Because success is what makes it expensive. That is a genuinely odd thing to budget for, and it is worth saying out loud to the person sponsoring the channel: if this works, here is what it costs, and that is a good problem which we should still see coming.
Second knowledge check. Your contact centre runs a hundred and forty agents for nine months and two hundred and twenty for three. What is the licensing risk? A, none, subscriptions flex with headcount. B, carrying peak headcount all year unless seasonal flex was negotiated in advance. C, paying overage during the peak months only. D, losing service levels because licences cannot be added quickly. Pause here before you continue.
B. Subscriptions commit annually, so without a negotiated mechanism the default outcome is that you buy for the peak and pay for eighty empty seats for three quarters of the year. A describes how people assume it works and it is not how the paper reads. C describes the outcome you would prefer, and it only exists if somebody negotiated it in advance. And D is a real operational worry that is downstream of the commercial one: the constraint is contractual, and it is entirely solvable ahead of time.
So, the shape of the year, five points. The peak is not a surprise: retail in November, travel in summer, finance at year end, and you know the month years ahead. A seasonal amendment gives you defined months and defined additional volume, agreed at the renewal when you have leverage. A committed base plus a flex band lets you buy the trough with a contractual right to add within a band at a held rate. Short term additions exist in some paper, so ask, and read carefully what you are given. And never negotiate in week one of peak. Let me put the arithmetic on this.
Guest analyst clip.
The only lever anybody has is your deadline. And notice this is the same principle as the renewal notice date from session two. Every one of these problems is solvable months in advance and nearly unsolvable in the week it arrives.
One more population question, because the service department is not uniform. Frontline agents work cases to a service level, so clearly Service Cloud and clearly the channels they use. Supervisors and workforce planners watch queues and rosters, usually the same licence and occasionally less. Escalation and back office specialists receive a handful of cases a week, which is worth testing rather than assuming. Quality and training people review recordings and transcripts, and are frequently over-licensed by default. And reporting and operations live in dashboards rather than the console, which is exactly the same conversation as session six's role map. Same exercise, different department, and it is worth doing with a service operations manager beside you.
Five failures. Double licensing on overlap, so both clouds issued because somebody occasionally looks at the other side. Channels bought for everybody, meaning the digital bundle applied to every agent when a third of them handle that channel. Peak carried all year, which is the single largest recurring waste in a seasonal contact centre. Voice consumption unwatched, priced per use, forecast once at purchase and never looked at again. And deflection treated purely as a service project, so its licence saving never appears in the business case that decides whether the project happens at all.
Last knowledge check. Which reduces the Service Cloud licence line most durably? A, negotiating a better per agent rate. B, reducing case volume, because agent count follows it. C, moving agents to Platform licences. D, consolidating channels into one bundle. Pause here and pick an answer before you continue.
B. Agent count is case volume divided by what an agent can handle, so anything that removes cases removes seats permanently, which is a structurally better outcome than a discount on seats you still need. A helps once and applies to the same quantity next year. C does not work for people genuinely handling cases and is the misuse session four warned about. D is worth doing and it optimises the price of a channel rather than the number of agents. Let me make the argument properly, because I think it belongs in your job rather than only in the service director's.
Guest analyst clip.
Often becomes obviously worth doing. So here is the calculation, and it is four steps. Work out cases per agent per year, which is total case volume divided by agent count, and almost nobody has that number to hand. Price one agent fully loaded, so licence plus channels plus add-ons, because that is what one avoided seat is worth annually. Convert deflection into seats, since ten percent fewer cases is roughly ten percent fewer agents at the same service level. Hand it to the service director, who is measured on satisfaction and handle time and has never been shown this figure. And revisit it at each renewal, because the ratio changes and the saving compounds if somebody keeps track.
Three sentences. Service Cloud carries the core sales objects, so the two cloud question is settled by whether the work on the other side is incidental or part of the job, and that single test resolves nearly every edge case your administrators will bring you. The core licence is a platform, and every channel beyond email and phone arrives priced on top of each agent who touches it, while voice behaves like consumption, which means success is what makes a channel expensive. And a seasonal contact centre carries peak headcount all year unless flex was negotiated in advance, while the most durable saving available is deflection, because agent count follows case volume and nobody has shown the service director what a seat costs. Next session, Platform licences in depth.
Homework before session eight, about two hours, and one number matters more than the rest. One, calculate cases per agent: annual case volume divided by agent headcount, and write it down because you will use it repeatedly. Two, price one fully loaded agent, licence plus channels plus add-ons, annually, because that is the value of one avoided seat. Three, list your channels and note which are per agent, which are consumption, and how many agents hold each. Four, chart your headcount by month, which is twelve numbers, and if that line is not flat you have a seasonal conversation to have at your next renewal. And five, find the overlap population holding both clouds, apply today's test, and see how many survive it.
Five guides, all on redresscompliance dot com. Salesforce Service Cloud licensing covers agents, supervisors, channels and the add-on layer, which is the reference version of today. Salesforce licence types explained is for settling the two cloud question and working through the overlap population. The Salesforce negotiation guide covers seasonal flex, committed bases and held rates, which is where the peak problem gets solved. Salesforce editions compared shows where the service tiers differ and what each step buys. And the Salesforce licensing assessment page describes what an independent review of an estate covers.
That is session seven. The thing to take away is that the test is the work rather than the org chart, channels get expensive when they succeed, the peak is predictable so negotiate it early, and deflection is the only lever that removes seats rather than repricing them. Next time, Platform licences in depth: what they genuinely cover, the limits, and the misuse patterns that create exposure. See you then.