Conversation Intelligence
AI Sales Call Analysis for B2B Services
Transcribes and analyses sales calls for talk ratio, next steps and risk signals, adapted for B2B Services businesses that run long, relationship-driven…
Conversation Intelligence
AI Sales Roleplay Trainer
Reps practise tough sales conversations with AI buyers built from your real calls, before they meet a live prospect.
New reps often learn your objections on real prospects, and managers rarely have time for regular roleplay. This trainer gives reps up to five AI buyers built from your own calls, ideal customer profile and objections. They practise by voice or text whenever they have time. Every session is scored against your sales method, and managers see a weekly scorecard.
- Buyer personas built from your own recorded calls
- Every practice session scored against your sales method
- Weekly manager scorecard in Slack, email or a sheet
$3,500
20–30 working days
Optional maintenance $249/mo
What it is
AI Sales Roleplay Trainer lets your reps practise sales calls by voice or text with up to five AI buyer personas. The personas are built from your recorded calls, ideal customer profile and objection list. Each session is scored against your own sales method, and managers get a weekly scorecard.
Why businesses buy it
A missed objection on a live call can cost a deal that took months to find. Classroom roleplay helps, but it depends on a manager's free time and a colleague willing to act as the buyer. This trainer makes practice available whenever a rep is ready, with buyers who sound like your market. Scoring follows the method your managers already coach to. The Scope Card fixes the number of personas, reps, languages and the scorecard destination before you order.
Problems this solves
New reps learn your objections on live prospects
Up to five AI buyer personas, built from your calls and objection list, let new reps practise those objections first. Each persona can be run by voice or text in a browser.
Managers have no time to run regular roleplays
Reps start practice sessions on their own, and every session is scored against your rubric. The weekly manager scorecard shows who practised and how scores moved.
Feedback on practice calls is inconsistent from one manager to another
The scoring rubric is built on your method, such as MEDDICC, SPIN or your own criteria. Every rep gets feedback against the same standard after each session.
A good fit if
New reps currently learn your pitch and objections on live prospects.
Managers know roleplay helps but rarely have time to run it every week.
You have recorded sales calls or written notes on the buyers your team meets.
You coach to a method such as MEDDICC or SPIN, or your own scoring criteria.
Not the right fit if
You want feedback on real calls after they happen, not practice beforehand. See AI Sales Call Analysis.
You want ongoing coaching notes for each rep from their live calls. See AI Sales Coaching.
You first need a clear objection list and suggested responses. See AI Objection Handling.
You need roleplays in a second language. Add the Additional Language Pack to the order.
How it works
- We study at least ten of your recorded calls or persona notes, plus your ideal customer profile and objection list.
- Up to five buyer personas are built, each with its own role, priorities and the objections your reps really hear.
- Your sales method is turned into a scoring rubric, such as MEDDICC, SPIN or your own criteria.
- Reps open a roleplay in their browser and practise by voice or text whenever they have time.
- After every session the rep gets feedback against the rubric, with what went well and what to try next.
- Each week the manager receives a scorecard in Slack, email or a sheet, showing practice and scores by rep.
What you provide within 72 hours
Your portal lists these the moment you order. Delivery starts the working day after everything is in and your access is checked.
- At least ten recorded sales calls, or written notes describing your typical buyers.
- Your ideal customer profile and your list of common objections.
- Your sales method or the criteria you use to judge a good call.
- The list of reps who will practise.
- A voice platform account in your name, shared by invite or a key you control, never by email.
- Your business information (company basics, hours and brand voice), shared during intake.
- One decision-maker who approves the personas and the scoring.
- An API key for your own AI model account (for example OpenAI or Google Gemini), created for this build with a spending limit.
The 72-hour rule. Send everything on your checklist, including access, within 72 hours of ordering so your build starts on time.
Timeline
Start
Intake verified; calls, objection list and sales method confirmed.
Weeks 1–2
Buyer personas and the scoring rubric built from your material.
Week 3
Voice and text roleplays set up, rep access and scorecard destination connected.
Weeks 3–4
Twenty practice sessions reviewed and summarized in a test report.
Week 4
Your review and two rounds of changes to personas and scoring.
Go-live
Reps invited, recorded handoff session, support starts.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $3,500 |
| Express | 15–25 working days | $4,100 |
| Rush | 10–20 working days | $4,750 |
What you receive at handover
- Up to five buyer persona roleplays, usable by voice and text
- A scoring rubric on your sales method, with feedback for each rep after every session
- A weekly manager scorecard in Slack, email or a sheet
- Setup for your reps in one language, within the Scope Card limit
- A test report on twenty practice sessions
- An admin guide, the recorded handoff session and 30 days of post-launch support
Expected benefits and how to measure them
- Can give new reps more practice before their first live calls
- Helps managers coach from scores rather than impressions
- Makes feedback on practice calls consistent across the team
- Practice sessions per rep each week
- Read the weekly scorecard after launch, and compare it with any roleplay sessions you logged before.
- Rubric score over time
- Compare each rep's average score in their first two weeks of practice with their score a month later.
- Time for a new rep to reach target activity
- Compare how long new hires took to reach your activity target before launch with hires who used the trainer.
Before you start: Record how often roleplays happen today, who runs them and how long new reps take to ramp.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
The build is paid once. Running it means voice platform minutes, billed per minute of practice by the provider, plus AI model usage. Both are billed to your own accounts.
Text roleplays use AI model usage only, so teams that practise mostly by text spend less on minutes.
Ask about running costsCustom requirements
More personas, more reps, an LMS or SSO connection, or certification tracking are outside this build. Add a note at checkout for a written Change Quote. Larger or bespoke projects go through the Custom AI enquiry form.
Request a custom versionKeep it running
After the 30 days of post-launch support, optional Upmark maintenance is available. It covers monitoring, fixes, platform updates and a monthly health report. Its change hours suit updates such as a new objection or a revised persona. Cancel any month.
$249/mo (the minimum fee, includes 2 change hours)
Optional. Starts after the post-launch support period. Cancel any month.
How maintenance worksHow this compares
| Do it yourself with a SaaS tool | Hire a custom agency | Upmark Labs | |
|---|---|---|---|
| Price | Roleplay software usually bills per rep each month | A sales trainer charges per workshop day | Ours is one fixed price, with no per-rep fee |
| Buyers | Generic personas from a template library | The trainer or a colleague plays the buyer | Ours builds personas from your own calls and objections |
| Scoring | The vendor's default rubric | The trainer's judgment on the day | Ours scores every session against your sales method |
| Availability | Whenever reps log in | Only on scheduled training days | Ours is open in a browser whenever a rep has time |
Questions about this product
How are the AI buyer personas built for our team?
From your own material. We use at least ten recorded calls or written persona notes, your ideal customer profile and your objection list. Each persona gets a role, priorities and the pushback your reps really hear. You review the personas and can ask for changes in two revision rounds.
Can reps practise by voice as well as text?
Yes. Every persona works by voice or text in a browser. Voice uses a voice platform account in your name, such as Vapi or Retell, billed per minute. Text practice uses AI model usage only. Reps can switch between the two depending on where they are.
Which sales methods can the scoring follow?
Your own. We can build the rubric on MEDDICC, SPIN or a similar method, or on criteria your managers already use. Each session is scored against it, and the rep sees what they covered and what they missed. The rubric can be adjusted during the two revision rounds.
What does a manager see each week from the roleplays?
A weekly scorecard in Slack, email or a sheet, whichever you choose at intake. It shows who practised, how often and how their scores moved against the rubric. Managers can use it to pick what to coach next, without sitting in on every practice session.
Does the roleplay trainer listen to our live sales calls?
No. It only runs practice sessions that reps start themselves. Live call monitoring and review are separate products, such as AI Sales Call Analysis. Your recorded calls are used only during setup, to shape the personas and objections, and they stay in your own storage.
What does voice roleplay practice cost to run each month?
It depends on how much your team practises. Voice platform minutes are billed per minute by the provider, plus AI model usage, all in your own accounts. We charge nothing per rep or per session. Text-only practice avoids voice minutes and costs less to run.
Can the trainer connect to our learning management system?
Not in this build. Reps use the roleplays in a browser, and results go to one scorecard destination. An LMS or SSO connection, or a certification program, is outside the Scope Card. Describe it in the custom requirements note at checkout and you get a written Change Quote.
Let reps make their mistakes in practice, not on live deals.
Order the AI Sales Roleplay Trainer and your reps practise with buyers built from your own calls. Managers get a weekly view of every score.
Works well with
Conversation Intelligence
AI Sales Call Analysis for Recruitment
Transcribes and analyses sales calls for talk ratio, next steps and risk signals, adapted for Recruitment businesses that want to coach consultants…
Conversation Intelligence
AI Sales Call Analysis for Automotive
Transcribes and analyses sales calls for talk ratio, next steps and risk signals, adapted for Automotive businesses that want to coach sales…