Best tools for practicing difficult conversations at work in 2026

EasyCoach is our top pick for practicing difficult conversations at work. Learning and development (L&D) teams build each roleplay from their own policies and playbooks and score it against their own criteria, at a flat fee with unlimited learners. Tough Tongue AI suits teams that want ready-made scenarios, Mursion suits high-stakes practice with a human option, and VirtualSpeech suits practice in virtual reality.

Key takeaways

• EasyCoach is our top pick for L&D teams that build their own practice. One platform covers feedback, underperformance, conflict, and bad news, from the same documents as the course, and the flat fee covers every learner.
• Counterpart realism no longer splits this field. All eight tools document counterparts that hold a position or push back, so realism alone will not narrow your shortlist.
• Who builds and scores the scenarios does. Most tools let your team build scenarios and set the criteria, while Mursion designs its own simulations and Skillsoft CAISY keeps custom building in a limited preview.
• Immersive options add avatars, headsets, and people. Mursion offers human-powered simulations for the highest stakes, and VirtualSpeech runs the same roleplays in the browser or in virtual reality (VR).

Difficult conversation practice lets people rehearse the moment before it happens

Difficult conversation practice is software that plays the other person in a hard workplace conversation, such as a performance review, a complaint, or a piece of bad news. The person practices out loud, the counterpart reacts and pushes back, and the tool scores the attempt when it ends.

Most people go into these conversations unprepared. In a March 2024 Achievers Workforce Institute survey of 1,500 US and Canadian employees, only one-quarter of managers had received enough training on hard conversations with direct reports, and one-third of employees said they feel unsafe having difficult discussions with their managers.

Practice only helps if it sounds like the real conversation. William Salm, Learning and Development Specialist at Equity Trust Company, builds his EasyCoach roleplays from what his best people say. “I’m stealing real language from my sales and support staff,” he says. “I want to know what my top representatives are saying. I want dialogue. I want the AI giving me feedback that’s very real.” (Equity Trust Company customer story). We explain the category in more depth in what is AI roleplay simulation? For practical steps, see how to use AI to prepare for a difficult conversation and our post on giving difficult feedback to a direct report.

Two kinds of tools sit close to this category but are not in this comparison.

• Courses and workshops on difficult conversations teach a framework, but they do not give people a counterpart to practice with on their own.
• General chatbots, such as ChatGPT or Microsoft Copilot, can play the other person in a chat. They have no consistent scorecard and no team reporting, so they work for preparation more than for practice. We cover this in can you use ChatGPT for coaching?

How we evaluated tools for practicing difficult conversations

We scored each tool on three criteria. The same three criteria run through the comparison table and the sub-headings inside every full entry below.

Counterpart realism. Whether the counterpart holds a position or pushes back, whether people practice by voice, and whether the scenario defines how the counterpart behaves.

Scenario setup and upkeep. Who builds the scenarios, from what material, and how much work it takes to update them when a policy or a process changes.

Feedback and scoring. Whether the tool scores each attempt against criteria you set, and whether L&D sees progress across a team.

How we graded each criterion

Counterpart realism. High means voice practice with a counterpart whose position, emotional state, or resistance the scenario defines, such as an employee who disputes a rating. Medium means voice or text practice with a counterpart that reacts to the learner, without documented settings for how it behaves. Low means text-only or scripted branching practice.

Scenario setup and upkeep. High means your own team builds and updates scenarios from its own material in a no-code tool. Medium means vendor-led setup, custom building in a limited preview, self-serve building the vendor does not document, or a reported rollout of two months or more. Low means the vendor builds every scenario.

Feedback and scoring. High means the tool scores each attempt against criteria you set, with team-level reporting. Medium means scoring on vendor-defined dimensions, or team reporting the vendor does not document. Low means no scoring.

Low is not a failing grade. It describes what a tool does, and a Low on one criterion can sit next to a High on another in the same entry.

Ownership disclosure

EasyCoach is our own product, and Easygenerator publishes this article. We reviewed all other tools independently, from G2 profiles and each vendor’s own published pages, and we checked all ratings and prices on November 2, 2026.

What we deliberately did not score

We did not test speech accuracy, response speed, or avatar quality ourselves, so run one of your own hardest conversations through each tool before you decide. We also did not score courses or workshops on difficult conversations, because this article compares practice tools. We did not grade pricing either, because the tools charge in ways that do not compare on one scale, from flat fees to per-seat plans, minutes, and content licenses. Each entry lists the pricing instead.

Comparison table of 8 tools for practicing difficult conversations

Counterpart realism no longer splits this field, because all eight tools document counterparts that push back. Who builds the scenarios and who sets the scoring criteria decide the shortlist instead.

We checked all G2 ratings and review counts in this table on November 2, 2026. G2 counts move in both directions as reviews are added and removed, so figures on pages published earlier may differ. “No rating confirmed” means we could not confirm a G2 score on that date.

Three kinds of tools for difficult conversations, and which one your team needs

Start with how high the stakes are and who will own the scenarios. If L&D builds practice for many teams, an AI practice platform fits best. If the conversation is high-stakes enough that the room, the body language, or a live person matters, an immersive simulation fits better. If your people already learn in Skillsoft or LinkedIn Learning, practice may already be part of your license.

• AI practice platforms your L&D team builds. These tools play whatever counterpart you describe, so one platform covers feedback, conflict, complaints, and bad news.
• Immersive and human-guided simulation. These tools add avatars, virtual rooms, VR headsets, or a live simulation specialist.
• Practice inside learning suites and AI coaches. These tools add roleplay to a content library or a coaching product your organization may already use.

AI practice platforms your L&D team builds

These tools play counterparts that your L&D team builds from its own policies and playbooks. We cover the wider category in our guide to the best AI roleplay tools for corporate training.

How the tools compare. EasyCoach is our top pick, for L&D teams that build the course and the matching roleplays in one platform at a flat fee. Tough Tongue AI is best for teams that want named difficult-conversation scenarios and published team plans. Yoodli is best for teams that want delivery coaching next to practice. Exec is best for teams that want structured programs with pass levels and certification.

Best fit for. Enterprises where many teams face their own hard conversations, from managers to customer support.

Tools in this category. EasyCoach, Tough Tongue AI, Yoodli, and Exec.

1. EasyCoach

EasyCoach is Easygenerator’s AI roleplay tool, sold standalone or bundled with an Easygenerator plan. Learners practice a hard conversation by voice or text against an AI counterpart that pushes back, hesitates, and changes the subject (EasyCoach).

EasyCoach sits inside a dedicated e-learning authoring platform. Your L&D team can build the course on giving feedback or handling complaints in Easygenerator, then build the matching roleplay from the same documents. At Odido, team managers practice difficult performance conversations and underperforming team members, while customer operations agents practice objections in their own scenarios (Odido customer story).

Counterpart realism

High. Authors define the counterpart’s role, background, goals, emotional state, and objections, and they can adjust its resistance level after the roleplay is generated (EasyCoach best practices). That lets an employee get defensive about a rating or a customer refuse a policy answer. Roleplays run in more than 70 languages, and a 2026 update made 40 of them newly voice enabled (Easygenerator product update). We found no documented support for several counterparts in one session.

Scenario setup and upkeep

High. Authors describe the conversation in a chat with the EasyCoach Assistant, which builds the roleplay. A requested change to the scenario, a role, or the tone updates the roleplay details, skills, and scoring criteria together (Easygenerator product update). Subject-matter experts can build scenarios too, not only instructional designers.

Feedback and scoring

High. EasyCoach scores every session against criteria your team defines and saves every attempt, so L&D sees progress across each team. Each organization decides whether managers see full sessions or only an aggregate skill score (EasyCoach).

Pricing. A flat fee with unlimited roleplays, sessions, and learners, so cost does not rise when more teams practice. Easygenerator does not publish the price, so contact sales for a quote (Easygenerator pricing).

Standout features

• Roleplays built from your own policies, playbooks, and real examples
• Counterparts with concerns, emotions, and a resistance level you set
• Scoring against criteria your team defines
• More than 70 languages, with voice practice in many of them
• Roleplays you can launch from your learning management system (LMS)

Pros

• One platform for managers, support, sales, and onboarding
• The course and the roleplay come from the same documents
• A flat fee with unlimited learners
• Customer stories from Odido and Equity Trust Company

Cons

• No prebuilt library of difficult-conversation scenarios
• No avatars or VR
• No published pricing
• Its G2 rating covers the whole Easygenerator platform, not EasyCoach alone

Best for

• L&D teams at enterprises with 1,000 or more employees that build their own practice
• Organizations where several teams face different hard conversations
• Multilingual teams

G2 rating. 4.7 out of 5 from 197 reviews, for the whole Easygenerator platform.

How we checked. Easygenerator’s G2 profile, the EasyCoach and pricing pages, the EasyCoach help center, product updates, and the Odido and Equity Trust Company customer stories, accessed November 2, 2026.

2. Tough Tongue AI

Tough Tongue AI is a platform for live AI roleplay agents, with named scenarios for the conversations people avoid, from performance reviews to pay and promotion talks (Tough Tongue AI).

Tough Tongue AI is best for teams that want difficult-conversation scenarios ready on day one. Its library includes a long-tenured employee who checked out after a harsh review and opens up only when the manager acknowledges it.

Counterpart realism

High. Each named scenario defines how the counterpart behaves, such as a top performer who expected a promotion and calls budget reasons moving the goalposts (Tough Tongue AI). People practice by voice in the browser, inside a Google Meet or Zoom call, or by phone.

Scenario setup and upkeep

High. In Scenario Studio, teams describe the situation and add their own values, competency frameworks, and case examples. Custom scenarios typically take 15 to 30 minutes to build (Tough Tongue AI).

Feedback and scoring

High. Every session ends with a report card that weights each skill in a rubric you write and scores it out of 10. Rules can cap a score, for example when someone confirms a colleague’s salary in a pay equity conversation (Tough Tongue AI).

Pricing. A free plan includes 25 one-time minutes, and paid plans start at $20 a month for 150 minutes. Team plans include Business at $299 a month for 15 seats and 2,990 minutes, and Momentum at $999 a month for 50 seats and 13,000 minutes (Tough Tongue AI pricing).

Standout features

• Named scenarios for reviews, promotions, pay, and re-engagement
• Weighted report cards with score caps you set
• Practice in the browser, in Google Meet or Zoom, or by phone

Pros

• You can start without a sales call
• Published team plans with pooled minutes
• Scenarios that reward specific behaviors, not only tone

Cons

• Usage-based minutes, so heavy practice raises the bill
• A newer vendor than most tools here
• We could not confirm a G2 rating

Best for

• Teams that want ready-made difficult-conversation scenarios
• Programs that want to write their own weighted rubric
• Small and mid-sized teams that want published prices

G2 rating. G2 lists a Tough Tongue AI profile, but we could not confirm a rating or review count on November 2, 2026.

How we checked. Tough Tongue AI’s leadership and pricing pages, accessed November 2, 2026.

3. Yoodli

Yoodli is an AI roleplay platform that grew out of speech coaching, with roleplays built on your company’s personas, objections, and scenarios (Yoodli on G2).

Yoodli is best for teams where how something is said matters as much as what is said. It measures pacing and filler words next to the content of the answer (Yoodli AI feedback).

Counterpart realism

High. Learners practice by voice with personas that admins configure, built on your company’s objections and scenarios, and the AI can present slides and documents during a roleplay (Yoodli on G2).

Scenario setup and upkeep

High. Business plans include Yoodli Create, which lets teams build their own roleplays and personas (Yoodli pricing).

Feedback and scoring

High. Every session is graded against the rubrics your team defines, and Measure dashboards show individual and team readiness (Yoodli on G2).

Pricing. The free Starter plan includes 5 lifetime sessions. Pro costs $8 a month and Advanced costs $20 a month, both billed annually, and team plans are priced per seat through sales (Yoodli pricing).

Standout features

• Delivery metrics next to roleplay feedback
• Custom rubrics and team dashboards on business plans
• LMS embedding through SCORM or LTI

Pros

• The lowest published individual price in this comparison
• You can try it without a sales call
• SOC 2 Type 2 and GDPR compliance

Cons

• Only 5 lifetime sessions on the free plan
• Team pricing is not published
• G2 reviewers mention analytics that could improve

Best for

• Teams that want practice and delivery coaching together
• People who get nervous and rush in hard conversations
• Pilots that start with a few learners

G2 rating. 4.8 out of 5 from 65 reviews.

How we checked. G2 profile and Yoodli’s pricing and AI feedback pages, accessed November 2, 2026.

4. Exec

Exec is an AI training platform for customer-facing teams and managers that combines AI roleplays, call scoring, courses, and certification programs (Exec on G2).

Exec is best for teams that want difficult conversations inside a structured program. Its manager guide includes scenario templates for difficult feedback, performance reviews, and saying no to a compensation request (Exec docs).

Counterpart realism

High. People practice by voice or with a video avatar, in scenarios with up to four AI characters (Exec on G2). Exec’s templates define how the counterpart reacts, such as a report who expected a better rating or who mentions an outside offer (Exec docs).

Scenario setup and upkeep

Medium. Scenario Studio builds a custom scenario in about two minutes from a plain-language description, your own content, or a recorded call. G2 users report an average implementation time of 2 months (Exec on G2).

Feedback and scoring

High. Each session ends with a scorecard, and programs can require a minimum score level before someone passes (Exec docs).

Pricing. A free plan includes 5 roleplay sessions per user per year. Starter costs $20 per seat per month with a 5-seat minimum, and Professional costs $39 per seat per month with a 15-seat minimum, both billed annually. Each seat includes monthly platform credits, and Enterprise is custom (Exec pricing).

Standout features

• Scenario templates for feedback, reviews, and compensation talks
• Minimum score levels and certifications
• Voice or video avatar roleplays with up to four AI characters

Pros

• 4.8 rating on G2
• You can start for free
• Courses and roleplay from one vendor

Cons

• Starter caps practice at 10 roleplay sessions per user per month
• G2 users report an average implementation time of 2 months
• G2 lists English as its only supported language

Best for

• Teams that certify people on difficult conversations
• Teams that want courses and roleplay from one vendor
• Buyers who want a published per-seat price

G2 rating. 4.8 out of 5 from 42 reviews.

How we checked. G2 profile, Exec’s manager training guide, and its pricing page, accessed November 2, 2026.

Immersive and human-guided simulation

These tools put the conversation in a virtual room, with avatars, body language, and in some cases a VR headset or a live simulation specialist.

How the tools compare. Mursion is best for high-stakes conversations where engineered resistance and a human-powered option matter. VirtualSpeech is best for teams that want a ready-made library of difficult conversations in the browser or in VR, plus custom roleplays their own team edits.

Best fit for. Programs where the setting, the avatar’s body language, or a live person changes how real the practice feels.

Tools in this category. Mursion and VirtualSpeech.

5. Mursion

Mursion is an immersive learning platform whose AI-powered simulations let employees rehearse tough conversations, such as performance reviews and change management (Mursion).

Mursion is best for the conversations that carry the most risk. It runs AI-powered simulations on demand and human-powered simulations for capstone and executive sessions (Mursion simulations).

Counterpart realism

High. Each scenario is built around a character with personality traits, and Mursion engineers the character’s resistance to the skill the scenario targets. During the conversation, AI sets the avatar’s emotional state in real time, such as defensive, receptive, frustrated, or open, and shows it through body language (Mursion).

Scenario setup and upkeep

Low. Mursion designs each simulation around its own research-backed skills framework (Mursion). We found no documentation of your own team building scenarios.

Feedback and scoring

Medium. Each skill in focus is rated on a scale from Developing to Strong, tied to observed behaviors and timestamped to the moment they happened, and the organization sees skill development at the cohort and program level (Mursion). The framework is Mursion’s, and we found no documentation that you set your own criteria.

Pricing. Mursion does not publish pricing, and access starts with a demo (Mursion).

Standout features

• Avatars whose emotional state changes in real time
• AI-powered and human-powered simulations
• Skill ratings timestamped to the moment in the conversation
• Areas from conflict and accountability to customer de-escalation

Pros

• Built for the hardest, highest-stakes conversations
• A human-powered option for capstone sessions
• Mursion says more than a third of Fortune 100 companies have used it

Cons

• Your team does not build its own scenarios
• No published pricing
• Only 2 G2 reviews

Best for

• Leadership programs with a high-stakes capstone
• Enterprises that want research-backed simulations without building them

G2 rating. 4.5 out of 5 from 2 reviews.

How we checked. G2 profile and Mursion’s homepage and simulation pages, accessed November 2, 2026.

6. VirtualSpeech

VirtualSpeech is a soft skills practice platform with AI roleplays in the browser, on a tablet, or in a VR headset (VirtualSpeech).

VirtualSpeech is best for teams that want ready-made difficult conversations on day one. Its library includes delivering layoff news, a poor performance review, disappointing promotion news, missed targets, and a disciplinary conversation about discriminatory language (VirtualSpeech).

Counterpart realism

High. Learners speak with an AI character who holds a position and pushes back, in a room that looks like the real one. If the learner handles it badly, the conversation gets harder (VirtualSpeech).

Scenario setup and upkeep

High. Teams describe the conversation, and VirtualSpeech drafts the character, their position, the situation, and a first set of scoring criteria. Your team then edits it like a document, with no code and no one from VirtualSpeech involved (VirtualSpeech custom roleplays).

Feedback and scoring

High. Every attempt is scored within seconds on five criteria chosen for that conversation, with evidence for each, and your team can edit those criteria (VirtualSpeech; VirtualSpeech custom roleplays). An AI coach then helps the learner reflect on the session (VirtualSpeech).

Pricing. Individuals pay $45 a month or $399 a year. Enterprise pricing depends on team size and rollout, and commissioned custom scenarios cost extra (VirtualSpeech pricing).

Standout features

• A ready-made library of difficult conversations
• Practice in the browser, on a tablet, or in VR
• Custom roleplays your team drafts and edits
• Five scoring criteria per scenario, with evidence

Pros

• Scenarios ready to assign on day one
• VR for conversations where the setting matters
• A published individual price

Cons

• No G2 reviews yet
• Enterprise pricing is not published
• Commissioned scenarios cost extra

Best for

• Teams that want a difficult-conversation library and VR in one tool
• Universities and enterprises that already use VR headsets

G2 rating. G2 lists a VirtualSpeech profile, but it had no reviews on November 2, 2026.

How we checked. G2 profile and VirtualSpeech’s roleplay, custom roleplay, difficult conversations, and pricing pages, accessed November 2, 2026.

Practice inside learning suites and AI coaches

These tools add roleplay to a content library or a coaching product your organization may already use, so people practice next to the courses or coaching they already have.

How the tools compare. Skillsoft CAISY is best for organizations that already license Skillsoft Percipio and want behavior profiles such as defensive or dismissive. LinkedIn Learning AI Role Play is best for organizations that already license LinkedIn Learning and want custom role plays from their own frameworks.

Best fit for. Organizations that want difficult-conversation practice without adding a new vendor.

Tools in this category. Skillsoft CAISY and LinkedIn Learning AI Role Play, with Valence, Coachello, Tenor, and CoachHub as short entries.

7. Skillsoft CAISY

Skillsoft CAISY is the AI conversation simulator inside Skillsoft Percipio. Learners choose a role, respond to the AI, and get adaptive, personalized feedback (Skillsoft documentation).

Skillsoft names coaching, conflict resolution, stakeholder negotiations, and HR discussions among CAISY’s uses. It fits organizations whose people already learn in Percipio.

Counterpart realism

High. CAISY uses behavior profiles that simulate how the other person responds, such as defensive, dismissive, or accepting (HCM Technology Report). Admins choose the default behavior, and learners can switch to another one or a random one before they start (Skillsoft AI settings). Learners type or speak, and can switch between voice and chat during a simulation (Skillsoft learner documentation).

Scenario setup and upkeep

Medium. Skillsoft supplies CAISY scenarios with its content licenses, and admins can create custom CAISY simulations in Percipio through a limited preview program, arranged with your Skillsoft account team (Skillsoft custom CAISY documentation).

Feedback and scoring

Medium. Each simulation ends with adaptive feedback and a detailed report, and learners see their best score and previous scores for each scenario (Skillsoft learner documentation). We found no documentation that you set your own scoring criteria.

Pricing. CAISY content is licensed with Skillsoft’s Expert collections or the Skillsoft Leadership Development Program (Skillsoft documentation).

Standout features

• Defensive, dismissive, accepting, or random behavior settings
• Voice and chat simulations inside Percipio
• A score history for each scenario

Pros

• Practice the same conversation against different behaviors
• No new vendor for Skillsoft customers

Cons

• Custom scenarios are still in limited preview
• Only available to Skillsoft customers
• Scoring follows Skillsoft’s model

Best for

• Organizations that already license Skillsoft Percipio
• Programs that want people to practice against a defensive or dismissive counterpart

G2 rating. 4.1 out of 5 from 107 reviews, for the whole Percipio platform.

How we checked. G2 profile, Skillsoft’s administrator and learner documentation, and HCM Technology Report’s launch coverage, accessed November 2, 2026. We go deeper in our EasyCoach vs. Skillsoft CAISY comparison and our Skillsoft CAISY alternatives guide.

8. LinkedIn Learning AI Role Play

LinkedIn Learning AI Role Play lets learners rehearse workplace conversations with an AI coach, such as handling difficult feedback or presenting to a skeptical stakeholder (Microsoft Support).

LinkedIn Learning AI Role Play fits organizations that already license LinkedIn Learning. The role plays also run in the Learning Agent in Microsoft Copilot.

Counterpart realism

High. The AI coach takes on the other party’s role, and each scenario sets out the goals and the roles and personality before the learner starts. Learners respond by voice or text (Microsoft Support).

Scenario setup and upkeep

High. Organizations on LinkedIn Learning Hub create role play scenarios tailored to their own priorities, using their own frameworks and best practices (LinkedIn Learning training). Admins can add the scenarios to learning paths and campaigns, and track how many conversations learners have.

Feedback and scoring

Medium. After each session, learners get strengths and areas for improvement tied to the transcript, plus course recommendations (Microsoft Support). We found no documentation that you set your own scoring criteria.

Pricing. AI Role Play comes with a LinkedIn Learning license.

Standout features

• Role plays in LinkedIn Learning and the Learning Agent in Microsoft Copilot
• Feedback tied to the transcript
• Course recommendations after each session

Pros

• No new vendor for LinkedIn Learning customers
• Custom role plays built from your own frameworks
• A large G2 review base for the platform

Cons

• Custom scenarios need LinkedIn Learning Hub
• Scoring follows LinkedIn Learning’s model
• Its G2 rating covers the whole LinkedIn Learning platform

Best for

• Organizations that already license LinkedIn Learning
• Teams that work in Microsoft Copilot

G2 rating. 4.4 out of 5 from 720 reviews, for the whole LinkedIn Learning platform.

How we checked. G2 profile, Microsoft Support’s LinkedIn Learning role play article, and LinkedIn’s training page on custom role plays, accessed November 2, 2026.

Short entries in this category

Valence. Valence offers Nadia, an AI coach whose role-play feature lets a manager practice a real upcoming conversation before it happens (Valence). Pricing is by request. Best for organizations that want ongoing coaching around hard conversations more than structured roleplay. We cover Valence in our guide to the best AI coaching software.

Coachello. Coachello combines AI avatar roleplays with AI coaching and ICF-certified human coaches, from giving feedback to managing conflict (Coachello on G2). It lists leadership coaching at €100 (about $112) per seat per month on G2, converted at the European Central Bank rate of October 30, 2026. Best for programs that want human coaches next to AI practice.

Tenor. Tenor offers more than 150 prebuilt leadership scenarios, from handling conflict to navigating change, in voice simulations with characters that react with emotion (Tenor in the ATD vendor directory). Pricing is by request. Best for leadership programs that want a large scenario library and a coach in Microsoft Teams and Slack.

CoachHub. CoachHub’s AI coach, AIMY, runs video role plays in which it takes the role of a peer, a manager, or a direct report and gives feedback afterward (CoachHub). Pricing is by request. Best for multinationals that want AI practice inside a human coaching program. We compare them directly in EasyCoach vs. CoachHub.

Which tool fits your difficult conversations

Pick by who needs to practice and how high the stakes are, then check how each tool scores it. This table names the tool we would shortlist first for each situation.

Where EasyCoach is not the right fit

EasyCoach is not the right choice if you need avatars, virtual rooms, or VR headsets, which Mursion and VirtualSpeech document. It also does not offer a ready-made library of difficult conversations, which Tough Tongue AI, VirtualSpeech, and Mursion provide, or a live human counterpart, which Mursion offers for its highest-stakes simulations. If your people already learn in Skillsoft or LinkedIn Learning, the roleplay in those suites adds no new vendor.

Test every shortlisted tool with your hardest real conversation

Take the difficult conversation your people avoid most, build it as a scenario in each tool, and have two people practice it. Watch whether the counterpart holds its position when they say something reasonable, and whether the score matches what your best manager would say about the attempt.