Certifications, without reading yourself to sleep
Pass your next certification with short questions and a tutor that has read the material
LEARN IT gathers the material for a module, writes exam-style questions from it, and measures how ready you are — domain by domain, at the official weightings. AB-250 is ready today, with more on the way. Everything lives on the platform. No twelve open tabs.
No lock-in. Tell us when you have passed.
In a new Authentication Settings record, what is the Public key URL used for?
Why it works
Because it is built around how people actually learn: a little at a time, with an answer to why — and a measure that does not congratulate you for being good at what you already knew. The same holds for each module as it arrives.
A module is the whole syllabus
Learn units and the official labs are pulled in and arranged by the exam’s own outline — headings, steps and warnings intact. The pipeline is the same for every new module, so they arrive in the shape the first one did.
Questions with their receipts
Every question is written from one specific unit and quotes the sentence the answer rests on, word for word. If the quote is not in the text, the question never ships.
A measure you can trust
Readiness is worked out per domain and weighted like the real paper. Get something wrong you used to get right and it counts down — otherwise you are only measuring yesterday.
Read aloud, at your pace
Every reading can be spoken aloud — start anywhere, pause, turn the speed up. It costs nothing and happens on your own machine.
Implement channels
Configure channel features
02 Message templates
Templates support the same dynamic fields as quick replies, so a representative can answer a common question without retyping it.
Quick replies
Short answers a representative can insert.
Sensitive data
Masks card numbers as they are typed.
Exam readiness
68%
Implement channels 30% · 74%
Configure work distribution 15% · 61%
Manage analytics 10% · 38%
Worth most next: Manage analytics
See a module from the inside
Real questions from one of the modules, and a tutor that answers from the unit the question was written from. No sign-up to try.
How the tutor answers
Which one is the right answer?
Not before you have committed — that part is the bit worth doing yourself. What I can do is make sure you have what you need.
Skill-based routing needs three things in place before it will assign anything: a rating model that defines the scale, the skills themselves attached to representatives at some proficiency on that scale, and a work classification ruleset that decides which skill a conversation demands. Miss any one and conversations fall through to the default queue.
So what does the rating model actually do?
It is the scale everything else is measured on — say 1 to 10, or Beginner to Expert. A skill on its own says a representative speaks Spanish; the rating model is what lets the system prefer somebody who speaks it at 9 over somebody at 4 when both are free.
“Skill-based routing requires a rating model that defines the scale used to measure agent proficiency.”
Source Configure routing rules and skill-based routing
The tutor answers from the unit the question was written from — and says so when the material does not cover it, rather than inventing something.
You need the Spanish skill attached whenever the customer language is Spanish - Mexico, Spanish - Puerto Rico, Spanish - Spain, or Spanish - United States. Why does the exercise have you build four separate rules instead of one?
Explanation
With conditions joined by And, a single rule listing all four values could never be satisfied, because one conversation has only one Customer Language. Splitting them means any one matching rule fires and attaches the Spanish skill.
Explanation
You can add multiple condition rows with **+ Add row**; the constraint is that they are combined with And rather than Or, which is what forces one rule per language variant.
Explanation
All four rules in this exercise share the same output - the Spanish skill at Conversational proficiency. The multiplication comes from the conditions, not the outputs.
Explanation
Order does matter in a decision list (which is why the Gold route-to-queue rule is moved to first position), but that is about priority between competing outcomes, not a reason to duplicate a rule per language variant.
In a real Contoso Coffee deployment, what do you do with the snippet copied from the Chat Widget Script section?
Explanation
The generated JavaScript snippet is what renders the widget on a page, and production deployment means placing it in the <head> of each support portal page. In this exercise the same snippet is instead saved into test.html to test end to end.
Explanation
The Behaviors page is where you handle Authentication settings, custom automated messages and the pre-conversation survey. It has no snippet upload, and the channel cannot publish itself onto your website.
Explanation
That section is the source of the snippet - you select Copy there. The widget is already active on the channel once Create channel completes; the snippet is only needed by the hosting web page.
Explanation
The workstream governs routing and work distribution for the conversations that arrive; the channel is linked to it on the Workstream details page. Serving the widget markup is the responsibility of your web pages.
In a production rollout modelled on this exercise, representatives complain they are receiving a fourth chat while three are still open, even though everyone holds the correct Spanish skill. Which configuration should you inspect first?
Explanation
Concurrency is governed by the work item limit, enforced only when Assignment blocking is Yes, and applied only to users listed on the profile's Users tab — the exercise's own note says a production deployment must assign the profile to all representatives handling chat.
Explanation
Proficiency affects how well a person matches skill-based routing criteria, not how many conversations they may hold simultaneously. The symptom described is a concurrency problem, so the capacity profile is the correct place to look.
Explanation
Persona mappings decide which workspace features and permissions a user gets. Removing or adding roles there would change what representatives can do, not throttle how much work arrives.
Explanation
If Type were wrong the skill could not be used for skill-based routing at all, but the scenario states skill matching is working. The overload is a capacity concern instead.
There are several hundred more inside the module — and a tutor to ask when one of them will not go in.
Get startedWhat the material looks like
A real extract, laid out the way you get it: headings, terms as cards, procedures as steps and Microsoft’s own warnings as warnings. Press the speaker and hear it — it happens on your own machine and costs nothing.
The phrase "AI-first contact center" is common in product descriptions. But what does it actually mean in practice, and how does Dynamics 365 Contact Center deliver on it?
A traditional contact center uses AI as an add-on: a chatbot here, a sentiment score there, perhaps an autosummarization feature bolted onto an existing agent desktop. An AI-first contact center is architecturally different. AI is embedded in every layer of the experience, from how customers get self-service answers to how agents handle interactions to how supervisors evaluate quality and plan staffing. It's not a separate system you add later. It's how the platform was designed from the start.
AI maturity in service: three levels
Microsoft describes the progression toward an AI-first contact center across three maturity levels:
| Level | Name | What it looks like |
|---|---|---|
| Level 1 | AI assistant | Every agent has Copilot available, helping them work faster and better. AI assists but doesn't act independently. |
| Level 2 | Digital colleagues | AI agents handle specific tasks alongside humans, such as discovering customer intents, drafting knowledge articles, or automating case closure. |
| Level 3 | Agent-operated | Humans set the strategy and direction; AI agents run entire service processes, checking in with humans at configured approval points. |
The extract stops here — the rest is inside the module. Read it on Microsoft Learn
Pricing
Pay for the module you are sitting, or get all of them. Paying yearly gives you two months free.
Prices in Danish kroner, excluding VAT. No lock-in — cancel whenever you like and access runs to the end of the period.
Get started
Leave your email and we will tell you when your module is ready — and give you a login now for the ones that already are.