{"id":11586,"date":"2026-10-28T12:14:00","date_gmt":"2026-10-28T03:14:00","guid":{"rendered":"https:\/\/blog.themusio.com\/?p=11586"},"modified":"2026-10-28T12:14:02","modified_gmt":"2026-10-28T03:14:02","slug":"how-a-pharma-sales-team-turned-a-low-score-into-a-follow-up-that-comes-back","status":"publish","type":"post","link":"https:\/\/blog.themusio.com\/?p=11586","title":{"rendered":"How a Pharma Sales Team Turned a Low Score into a Follow-Up That Comes Back"},"content":{"rendered":"\n

A composite scenario. This story is illustrative, built from how EOS actually works. It isn’t a report on a specific customer, and its quotes are illustrative, not verbatim.<\/em><\/p>\n\n\n\n

A sales rep finishes a practice session. Once it’s evaluated, he sees a product-knowledge score below 80.<\/p>\n\n\n\n

That score doesn’t alert his manager. It simply shows up in his results, the way a low score always does.<\/p>\n\n\n\n

But something else happens in the background, quietly and automatically, without anyone asking for it. EOS generates a remedial quiz built from the exact claims he got wrong in that session. It isn’t a generic refresher on the product line. It’s a quiz made of his own mistakes.<\/p>\n\n\n\n

He finds it in his own quiz area.<\/p>\n\n\n\n

This is the story of how a pharmaceutical sales organization could use remedial quizzes and spaced review to follow up on knowledge gaps that would otherwise stay invisible, and what it would learn about the limits of that loop.<\/p>\n\n\n\n

The setup: a product line where approximate knowledge is a liability<\/strong><\/h2>\n\n\n\n

The company sells prescription products to physicians and specialists through a field force of territory representatives. The portfolio is substantial: multiple indications, dosing ranges that vary by patient population, contraindications, interaction warnings, and payer-coverage conditions that shift with formulary changes.<\/p>\n\n\n\n

A rep who gets a dosing limit approximately right isn’t approximately safe. They’re wrong, and in a regulated environment a wrong claim isn’t just a coaching opportunity.<\/p>\n\n\n\n

The company uses EOS for practice sessions and claim-level verification. Reps practice against AI personas, and once a session is evaluated, every factual claim they made is checked against the company’s approved product knowledge. When a claim comes back Incorrect or Partially Correct, the rep sees the verdict and the reasoning behind it.<\/p>\n\n\n\n

But knowing in the moment isn’t the same as remembering next month. That’s the gap this company hadn’t addressed.<\/p>\n\n\n\n

The problem: a correction that doesn’t carry over<\/strong><\/h2>\n\n\n\n

The training team noticed a pattern. Reps who scored below 80 would read the feedback on their missed claims and, in the moment, understand what they’d gotten wrong. Weeks later, some of them made the same kind of mistake again.<\/p>\n\n\n\n

The team could only find that by reading session reports. EOS keeps every attempt, but it doesn’t count how many times a rep has missed the same question or claim, so spotting a pattern meant looking.<\/p>\n\n\n\n

“We were catching the error in practice, showing the rep what was wrong, and moving on. The rep understood it that day. But understanding something once and being able to recall it weeks later are two different things. We were only solving the first one.”<\/em><\/p>\n\n\n\n

What was already running<\/strong><\/h2>\n\n\n\n

Nothing needed to be adopted. When a session’s product-knowledge score came in below 80, EOS had been generating a quiz in the background all along, and scheduling the questions to come back. The team simply hadn’t been paying attention to it. Here’s how the loop works, and what the team noticed at each step.<\/p>\n\n\n\n

Step 1: The trigger.<\/strong> When a session’s product-knowledge score is below 80, EOS generates a quiz in the background, built from the claims the rep got wrong. If there aren’t enough failed claims to fill a quiz, it adds extra questions on the same product. The questions can be multiple choice or free response. Nobody has to ask for it. The rep finds it in their own quiz area, and managers aren’t notified.<\/p>\n\n\n\n

Step 2: The rep’s first encounter.<\/strong> On the Quiz page, the rep sees two paths under “Start a quiz”: New quiz, and Review, labeled “Revisit the questions you missed.” The home dashboard also shows a small chip with the number of reviews due. The rep sees a count, not a list. The Review panel doesn’t show the text of the due questions, so the rep finds out what’s due by starting a review. The Quiz page also shows their own stats: attempts, accuracy, a streak counter, and focus areas, which are the products where their accuracy is lowest, each with a Practice button.<\/p>\n\n\n\n

Step 3: The review schedule.<\/strong> Every question a rep answers gets its own review date, for that rep. A question nobody has tried has no date. Intervals grow as the rep keeps answering correctly, and a miss brings the next review back to one day. The schedule belongs to the individual question and the individual rep, so two reps on the same question each have their own date.<\/p>\n\n\n\n

Step 4: What the rep sees after each answer.<\/strong> Right after submitting, the rep sees Correct or Incorrect and a short note on why. Free-response answers are graded by a model against that question’s rubric, and the rep can see the rubric afterward. If the question has an explanation, it shows too, along with a source name and a link when one was kept. Two caveats the team learned early: the link doesn’t always point to the exact page a fact came from, and if grading itself fails, the attempt is marked Incorrect with an error message instead of a real explanation, so an unexpectedly harsh result is worth a second look. Admins can override a grade.<\/p>\n\n\n\n

What the team notices in the first weeks<\/strong><\/h2>\n\n\n\n

Three things stand out, and not all of them are flattering.<\/p>\n\n\n\n

The loop runs quietly, including the backlog.<\/strong> EOS doesn’t send a reminder when a review comes due. A rep sees the due count only when they open the Quiz page or their home dashboard. Reps who open those screens see the count and start reviews. Reps who don’t simply accumulate due questions, because overdue reviews never expire. A review sitting also serves only a limited batch, so a large backlog takes more than one sitting. The training team’s fix is behavioral, not technical: tell every rep where a due review shows up.<\/p>\n\n\n\n

The people with the visibility are org admins and owners.<\/strong> Trainers and other roles don’t see the team-wide view at all. The enablement lead, an org admin, opens the organization stats, which show one team-wide number for how many questions the team needs to revisit. From there she opens team attempts, filters by trainee and product, and reads individual attempts: the question, the rep’s answer, the verdict, the feedback, and the next review date. She also uses the Curriculum Gap heatmap and the needs-attention list, which flags reps who have gone quiet for about a week, or whose scores or quiz accuracy have dropped. She turns that into a weekly check-in and brings specifics to one-on-ones.<\/p>\n\n\n\n

The loop shows activity, not resolution.<\/strong> There’s no “gap closed” or “mastered” status. A run of correct answers just pushes that question’s next review further out. Nothing counts how many times a rep missed the same question, and a rep with steady low accuracy and no recent change never appears on the needs-attention list. There’s also no list of who is late on what. The team learns to treat the loop as a prompt, not as proof.<\/p>\n\n\n\n

\n

Does your training loop end at the score?<\/strong> Our 12-page report, Beyond Roleplay: The Rise of the Sales Knowledge Engine<\/em>, covers claim extraction, fact-grounded verification, and the questions worth asking any vendor about how knowledge gaps are found and followed up. [Download the whitepaper \u2192]<\/strong><\/a><\/p>\n<\/blockquote>\n\n\n\n

Two things that need an admin’s attention<\/strong><\/h2>\n\n\n\n

When a product changes.<\/strong> Editing an approved product refreshes the knowledge used for later scoring and newly generated questions, but it doesn’t rewrite, deactivate, or delete quiz questions already in the system, and it doesn’t clear scheduled reviews. After a product label changes, a rep can still be quizzed on something written from the older label. An admin has to deactivate the outdated questions by hand. That becomes a standing step whenever a product changes.<\/p>\n\n\n\n

When a session is evaluated again.<\/strong> Running Evaluate again replaces the product-knowledge result, including the claim verdicts, and the report doesn’t mark the replacement. If the new score is below 80, the earlier remedial quiz is deleted, along with the attempts on it, and a new one is generated. If the new score is 80 or above, the earlier questions stay, even though the claims they were written from have been replaced. The team adopts a habit: evaluate a session once, and re-run it only when there’s a clear reason.<\/p>\n\n\n\n

“The remedial quiz didn’t replace coaching. It gave coaching something specific to point at. Instead of saying ‘review the product guide,’ I could open someone’s attempts and say ‘here are the interaction questions you missed on Tuesday, let’s walk through them.’ That’s a different conversation. It just isn’t proof anyone has fixed anything.”<\/em><\/p>\n\n\n\n

Why this loop matters in regulated sales<\/strong><\/h2>\n\n\n\n

Pharmaceutical sales sits at the intersection of two problems: a product catalog where approximate knowledge can create compliance risk, and a field force that is distributed, busy, and hard to retrain at scale.<\/p>\n\n\n\n

The company already had claim-level verification. They knew when a rep got something wrong. What they added was attention to a follow-up that comes back on its own schedule, with no manager effort. Verification caught the error. The quiz built a targeted response from it. The review schedule brought that question back.<\/p>\n\n\n\n

What the loop does not do is confirm that a gap closed, alert anyone to a stubborn one, or keep up with the catalog without help from an admin. And EOS checks what a rep says against the product knowledge the company has approved in EOS. It isn’t a medical-legal review tool and doesn’t replace that process. For a compliance-sensitive team, knowing exactly where those lines sit is part of using it well.<\/p>\n\n\n\n

The takeaway for sales leaders<\/strong><\/h2>\n\n\n\n

If your reps get feedback on their mistakes but make similar mistakes weeks later, you have a follow-up problem as well as a verification problem.<\/p>\n\n\n\n

Catching an error in practice is necessary. It isn’t sufficient. The rep who understands a correction on Tuesday may not recall it weeks later.<\/p>\n\n\n\n

A remedial quiz built from a rep’s own failed claims, paired with a review schedule that brings those questions back, gives the correction a chance to come back at all. Pair it with a few habits: tell reps where due reviews show up, look at actual attempts rather than just counts, deactivate old questions when a product changes, and think before re-running an evaluation.<\/p>\n\n\n\n

That’s how a low score stops being the end of a conversation and becomes the start of a follow-up.<\/p>\n\n\n\n

\n

A low score should start a follow-up, not end a conversation.<\/strong> EOS builds remedial quizzes from a rep’s own failed claims and schedules them for spaced review. Start free with up to 5 seats at app.akaeos.com<\/strong>, or [download the full whitepaper]<\/strong><\/a>.<\/p>\n<\/blockquote>\n","protected":false},"excerpt":{"rendered":"

A rep mixes up a product claim in practice. EOS builds a quiz from it and schedules it to return. A composite scenario of what the loop does, and doesn’t.<\/p>\n","protected":false},"author":1,"featured_media":11587,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3642,3640],"tags":[4723,4946,5030,5137,5135,5024,4724,5026,5025,5028],"class_list":["post-11586","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-en","category-all-en","tag-ai-sales-training","tag-case-study","tag-knowledge-gaps","tag-pharma-sales","tag-pharmaceutical-sales-training","tag-remedial-quiz","tag-sales-enablement","tag-sales-training-quiz","tag-spaced-repetition","tag-spaced-review"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t