The metrics that actually matter for your academy

Ask ten academy owners which numbers they watch and you get almost the same three answers: enrolments, monthly revenue, courses published. All three are accurate. Not one of them tells you what to do on Monday morning.

That is the measurement problem in most academies. We track what is easy to count instead of what changes a decision. A number earns a place in your weekly review on one condition: there has to be a reading of it that makes you act differently than you would have without it. Everything else is dashboard decoration. This article covers how to filter your numbers with that test, which questions deserve a metric at all, where each number actually lives inside your academy, and what the data will never tell you no matter how much of it you collect.

A number that changes no decision is not a metric

Take enrolments. They went from forty to fifty-two this month. Good news, and now what? Nothing specific. The number tells you something worked without saying where it worked or how to repeat it. Now take a second question: how many of this month's buyers opened lesson one and never came back? That number has one obvious response, and it is a message you can write today to people whose names you know.

The difference between the two is not accuracy, because both are accurate. The difference is what follows them. An outcome number measures a result that accumulated from ten causes at once, so it points at none of them. A diagnostic number measures a single step in the learner's path, so it points at what sits on either side of that step. Watching outcomes alone is why the familiar feeling shows up: plenty of numbers, no idea what to do with any of them.

So run a test before you add anything to your review. Next to the number, write what you would do if it went up, and what you would do if it went down. If the two answers look alike, or if neither comes easily, the number has not earned a weekly slot. Send it to the annual report. If the answers genuinely differ, you have found a metric. This one test deletes about half of what most people track, which is a gain rather than a loss: five numbers you read properly beat thirty you glance past.

Five questions worth a number, and everything else can wait

A learner moves through five steps with you. They arrive, they buy, they start, they understand, and they hear from you again. Each step gets one question and one number, which is enough to begin with. The table below ties each question to where the number lives and to the decision it produces, because that last column is what turns a figure into a metric.

Question	The number that answers it	Where you read it	The decision that follows
Are people arriving at all?	Public page traffic and its source	The tracking tools you connect	Repeat the channel that brought them, or drop it
Are they buying?	Sales, and which product they came from	Sales and wallet reports	Adjust price, sales page, or how the offer is ordered
Do they start after buying?	Progress for each learner separately	The learner's own page in the learners area	Write to the people who never started, not to everyone
Do they understand?	Who passed and who needs a retake	The report inside each quiz or assignment	Re-teach the idea, or rewrite the question
Do your emails land?	Delivery log, opens and engagement	Email analytics, next to the emails	Change the subject line, the timing, or the segment

Row three is the one most often ignored and the cheapest to fix. Someone who paid and then opened nothing is not a lost customer yet. They are a busy one, and what separates them from someone who has given up is that they still want the thing. A single message written to them alone reads differently from a campaign blasted at every buyer, because it knows their situation. You cannot target like that without a number telling you exactly who they are.

Notice what the table does not contain: any comparison against an industry figure. The only comparison that helps you is against your own previous cohort, on the same product at the same price. Published averages are pooled from audiences, price points and sales journeys that look nothing like yours.

Every number sits next to the thing it describes

New academy owners go looking for a page called Reports that gathers everything in one place. Dwrrah does not have one, and the reason is practical rather than a gap: a number is useful beside the thing it describes, because the moment you read it you are already on the screen where you would act. A quiz report inside the quiz means you rewrite the question while looking at its results, instead of opening three pages to connect them.

Everyone starts from the same place. The analytics dashboard is on every plan at no extra cost and shows your academy's movement in a single view, enrolments, sales and general activity, so you know where to look before opening anything detailed. From there you go down to where the detail lives:

• Wallet and sales reports on the financial side, showing exactly which product your sales came from.
• Email analytics beside the emails themselves, with a delivery log recording what left, when, and to whom.
• A standalone report inside every quiz and every assignment.
• Per-learner progress from their record in the learners area.
• Export as CSV or JSON when you want to pull the numbers into your own reporting or analyse them in the tool you prefer.

The Pro plan adds three layers on top. Content performance shows how learners move through your lessons and where they stop. The activity log records what happened inside your academy and who did it, which is the thing you need the day a team starts working alongside you. And tracking pixels cover tools such as GA4, Meta, TikTok and PostHog. Those pixels inject only on your public academy pages, the part a visitor sees, and never run inside the dashboard.

The assessment report splits four states, not one

The most misused number in teaching is the pass rate. It tells you sixty percent passed and says nothing about the other forty, who were never a single group to begin with. This is why the report inside each quiz and assignment separates four states: who was entitled, who started, who passed, and who needs a retake.

That split keeps two unrelated problems apart. A learner the assessment never reached has an entitlement or sequencing problem, and the fix is administrative, with nothing to do with the quality of your questions. A learner who sat the assessment and failed has a comprehension problem, or a problem with the question itself, and the fix lives in the material or the wording. Blend them into one percentage and you get the wrong move in both directions: re-teaching a lesson to people who never reached it, or sending administrative follow-ups to people who understood the lesson and tripped over an ambiguous question.

Which leads to the most useful question to ask after any assessment: is the question everyone failed hard, or is it unclear? If your strongest and weakest learners fail it at roughly the same rate, it is almost certainly unclear, and the repair belongs in the wording rather than in your teaching.

What the numbers will not tell you

This section is what makes the rest of the article believable.

• Analytics present data, they do not create it. Progress comes from lessons being watched, revenue from purchases, assessment figures from attempts, email numbers from the delivery log. The less of the platform you use, the thinner the picture, because reporting cannot recover what never happened inside it.
• Data tells you where, not why. You will learn that five people stopped at lesson seven and nothing about the reason. Ask three of them directly. One honest reply teaches you more than a full dashboard.
• There is no universal benchmark to measure yourself against. We will not hand you a normal completion or conversion rate, because it moves with price, length, audience and how the course was sold. Anyone who gives you that figure is selling reassurance, not information.
• There are no learner points, badges or leaderboards on the platform. If you are shopping for a composite engagement score of that kind, you will not find it here. The points named elsewhere in Dwrrah are a billing unit rather than a reward: live-stream points meter attendance by the hour, and creation credits meter what Allam generates for you. What gets measured on your side is progress, assessment results and recorded attendance.
• A number repairs nothing on its own. A dashboard you open every morning that produces no action is a comfortable habit, not work. The value sits entirely in the last column of the table.

Five measurement mistakes, then a one-month routine

First: watching revenue alone. Revenue is a lagging result that reports on last month, and by the time it dips the cause has already finished happening a step earlier. Pair it with at least one early number from the first three rows of the table.

Second: reading the numbers daily. Daily movement is noise, and reading noise pushes you into rushed changes and then into reversing them. Read weekly, decide monthly, and never judge a change before it has had two weeks.

Third: changing two things at once. Raise the price and rewrite the sales page on the same day and no result will be attributable to either. You will have spent a month to earn one ambiguous lesson.

Fourth: measuring what you have no intention of changing. Some numbers get tracked because they exist, not because they help. Put them through the up-or-down test and delete whatever fails it.

Fifth: stopping at the data instead of asking a human. The best report gives you a hypothesis, and a short conversation with one learner confirms or kills it in ten minutes. Make the direct question part of the routine rather than an emergency measure.

A one-month routine to start with

1. Pick three numbers from the table and write the decision each one drives. Three is enough for month one.
2. Record today's values as a baseline. Without one you will never know later whether things improved or you simply got used to the figure.
3. Read them once a week at a fixed time, and change exactly one thing in response.
4. Open your most recent assessment report and separate the people it never reached from the people who failed it. Treat each group its own way.
5. At month end, ask three learners who stopped why they stopped, and compare their answer with what you assumed from the data.

Basic is 99 SAR a month, Pro is 199, Premium is 299, and the platform takes 0% commission on your sales, with everything laid out on the plans page. Start with the free dashboard on one course you already run through a 14-day free trial with no card, and pick three numbers you already know what to do with.