Who’s With You — Quick Introductions

• Who from your team is joining this conversation (pick all that apply)? Options: VP of Design / Head of Engineering, CAD Manager / Lead, RTL Architect, Block Owner / IP Lead, Physical Design / Floorplanner, Verification Lead, Other
• Who is the primary decision owner for front-end tool selection on this project? Options: VP of Design, CAD Manager, Director of Architecture, Procurement/Tooling Manager, Program Manager, Other
• What is the approximate size of the team that will interact with the synthesis flow (CAD + RTL + P&R touchpoints)? Options: < 5 engineers, 5–15 engineers, 16–50 engineers, 51–200 engineers, > 200 engineers
• How soon is your next tapeout or major milestone that could be affected by front-end QoR? Options: Within 1 month, 1–3 months, 3–6 months, 6–12 months, More than 12 months
• Briefly describe a recent example where synthesis or ECO work impacted schedule (what happened and how long it took).

What Keeps Your Tapeouts Awake at Night?

• Could your current synthesis flow be silently adding weeks to every tapeout? Options: Yes — regularly, Sometimes, Rarely, Not sure
• How often do critical blocks require manual ECO iterations after placement to meet timing? Options: Almost every block, Several blocks per project, Occasional problem blocks, Rarely
• When those ECO loops happen, what’s the typical real cost—calendar time and engineering hours? Options: < 1 week / < 40 engineer-hours, 1–2 weeks / 40–160 engineer-hours, 2–4 weeks / 160–400 engineer-hours, > 4 weeks / > 400 engineer-hours
• How does that experience feel for your team—frustrating, exhausting, accepted as normal, or something else? Options: Frustrating and urgent, Exhausting but manageable, Accepted as normal, Demoralizing / hurts retention, Other
• Which downstream impact worries you most when front-end QoR misses target? Options: Schedule delay, Increased silicon respins, Budget overrun, Quality/regression risk, Loss of market window, Other

The Real Hard Blocks — Show Me the Scars

• Which single IP block do you most dread giving to synthesis, and why?
• What are the concrete failure modes you see on those blocks? (select all that apply) Options: Timing violations after P&R, Excessive congestion post-place, Cross-domain CDC failures, Linting or stylistic violations preventing integration, Runtime blowups (memory/swap), Other
• What characteristics make those blocks hard (pick all that apply)? Options: High fanout nets, Complex clocking / many domains, Large sequential depth, Heavy use of macros or hard IP, Asynchronous interfaces, Aggressive physical constraints (floorplan), Other
• For a typical hard block, what are its scale metrics (approx)? Options: < 10k cells, 10k–50k cells, 50k–200k cells, > 200k cells
• How do you currently prioritize which blocks to optimize when timing fails—by risk to tapeout, power, or other criteria? Options: Risk to tapeout / critical path, Power impact, Area impact, Customer-visible features, Other

What Would Winning on This Project Really Mean?

• If a new synthesis flow delivered measurable improvements, what single outcome would convince you to change tooling? Options: Timing closure on hardest blocks, Overall PPA improvement, Faster turnaround/runtime, Reduced manual ECOs, Easier migration / lower requalification cost, Other
• What specific timing/QoR targets would you consider a success for pilot blocks (e.g., % slack improvement or reduced iterations)? Options: < 5% QoR improvement, 5–10% improvement, 10–20% improvement, > 20% improvement, Unsure—need help defining
• What runtime or throughput expectations do you have for synthesis on large blocks or full-chip runs? Options: Similar to current tool, Up to 2× slower acceptable if QoR improves, Must be faster than current tool, Depends on batch scheduling and cloud usage, Unsure
• How much requalification effort (person-weeks, regression scope) would be acceptable to adopt a new front-end tool? Options: < 2 person-weeks, 2–8 person-weeks, 2–3 person-months, > 3 person-months, Unsure / depends on value
• Beyond metrics, what non-technical signals would make you confident (e.g., vendor support, migration scripts, reference designs)? Options: Strong CAD support & training, Migration scripts and adapters, Customer reference case studies, Formal verification artifacts, On-site pilot collaboration, Other

Assumptions That Could Be Killing Your Schedule

• Which assumptions about your current flow could be wrong—and what would change if they are?
• How tightly coupled are your scripts & flows to your incumbent synthesis tool? Options: Loosely — modular scripts, Moderately — some dependencies, Tightly — heavy vendor-specific scripts, Don’t know
• Which migration tasks worry you most (select all that apply)? Options: Script translation, IP/legal requalification, P&R tool interoperability, Regression verification, CI/CD integration, Other
• Do you have internal automation or CI that would need to be adapted? If yes, describe the scope. Options: No automation, Light automation (per-block), Full CI/CD for flows, Custom in-house tools (extensive)
• If migration fails, what is your rollback plan and how long would rollback take? Options: Immediate rollback, minimal work, Rollback within weeks, Rollback months — significant work, No clear rollback plan

Risk Appetite: How Much Change Can You Tolerate?

• How much migration risk would you accept in exchange for a 10–20% QoR gain on critical blocks? Options: High — willing to requalify, Moderate — limited requalification, Low — tidy migration only, Zero — cannot requalify
• What governance or executive approvals are required to pilot or adopt a new front-end tool? Options: VP-level approval, CAD Manager approval, Architecture committee, Procurement / legal, Board / exec steering, Other
• What compliance, IP, or export constraints could limit a cloud-based pilot or vendor collaboration? Options: Export controls (e.g., EAR/ITAR), IP transfer limits, On-prem only requirement, No significant constraints, Unsure—need to check
• What does an acceptable rollback/escape clause look like to your legal/procurement team? Options: Timeboxed pilot with defined gates, Money-back / limited liability, Limited license term, Co-development terms, Other
• How will success/failure of a pilot be communicated internally (who needs updates and cadence)? Options: Weekly CAD syncs, Biweekly exec reviews, Formal milestone reports, Ad-hoc as needed, Other

The Hands & Process: Who Will Do the Work?

• Who will own day-to-day migration tasks and pilot runs on your side? Options: Dedicated CAD engineer(s), Block owners with CAD support, Shared across teams, External consultants, Other
• How many full-time equivalent CAD engineers could realistically be allocated to a pilot over a 3-month period? Options: < 0.5 FTE, 0.5–1 FTE, 1–3 FTEs, 3–6 FTEs, > 6 FTEs
• What training formats are most effective for your CAD/RTL teams (pick all that apply)? Options: Hands-on paired sessions, Week-long workshop, Recorded tutorials & docs, On-site mentoring during pilot, Regular office hours / support
• Which internal owners must sign off on acceptance criteria (e.g., CAD manager, block owner, verification)? Options: CAD Manager, Block/IP Owner, Verification Lead, Physical Design Lead, Program Manager, Other
• Describe any past migrations or tool adoptions your team ran—what went well and what surprised you?

Pilot Design — Be Surgical, Not Scattershot

• If we ran one pilot that would be indisputable evidence of value, what must it include?
• Which benchmark blocks would you pick for a pilot (select up to three)? Options: Most timing-critical block, Largest block by cell count, Most congested block, Cross-clock-domain block, Customer-visible feature block, Other
• Which deliverables would you require from the pilot to accept results (select all that apply)? Options: Detailed QoR report (timing/power/area), PPA comparison scripts, Reproducible runbooks & scripts, Migration checklist, Support plan for requalification, Other
• What acceptance criteria would constitute a pass for each pilot block (be specific: slack improvement, iterations saved, runtime threshold)?
• What minimal timeline would you accept for a pilot from kickoff to acceptance? Options: 2–4 weeks, 4–8 weeks, 8–12 weeks, > 12 weeks

Decision Triggers: What Will Make You Say Yes?

• What single metric would make your VP of Design approve a roll-out (pick one)? Options: % timing closure improvement, % fewer ECO iterations, Net PPA uplift, Reduction in engineer-hours, Faster turn-around time, Other
• What procurement or budget windows are relevant—when would you need commercial terms in place? Options: Immediately, Next quarter, Next fiscal year, Budget TBD, Unsure
• What remaining risks or objections, if any, would block a pilot from moving forward? Options: Technical risk, Migration cost, Procurement/legal, Staff bandwidth, Security/compliance, Other
• If we propose a pilot, who else should we involve in the kickoff to make it effective (names/titles)?
• What next step would you prefer after this discovery call (pick one)? Options: Technical deep-dive workshop, Pilot proposal & SOW, Reference customer intro, Internal alignment meeting, Other

Start Here: A Quick Snapshot of Today

• Which phase best describes where your project is right now? Options: Early RTL development, Synthesis & optimization, Timing closure on IP blocks, Full-chip P&R, Pre-tapeout validation, Other
• Who is the primary buyer or stakeholder we should align with for discovery (role/title)? Options: VP Design/Engineering, CAD Manager/Lead, Block Owner / Principal Designer, RTL Architect, PM / Program Manager, Other
• Briefly describe your current synthesis/flow headline — one sentence that captures what’s working and what isn’t.
• Which metrics do you actively track during synthesis and early signoff? Options: WNS/TNS, Setup/Totals of hold violations, Area (gate count), Estimated power, Turnaround/runtime, Lint/CDC violations, Other
• What size are the blocks you most frequently run through synthesis? Options: Small (<10K regs), Medium (10K–100K regs), Large (100K–1M regs), Very large (>1M regs), Varies widely

Why Do We Still Lose Time to ECO Loops?

• When you look at your last three tapeout delays, how often did manual RTL→ECO loops cause the slip? Options: In every delay, Most of the time, Occasionally, Rarely, Never
• How many RTL-to-layout ECO iterations does a typical hard block need before timing converges (pick the usual case)? Options: 0–1 iterations, 2–3 iterations, 4–6 iterations, More than 6
• Which root causes most often force those ECO loops? Options: Place-and-route surprises, Clocking/Cross-domain issues, Unexpected congestion, Synthesis pessimation/optimization limits, Constraint/spec mismatches, Tool integration bugs, Other
• How long does a single ECO loop typically take from identification to validated regression (calendar time)? Options: Hours, 1–3 days, 4–7 days, 1–2 weeks, More than 2 weeks
• Describe one recent ECO loop that cost the team time — what happened and what felt most painful?

Where The Synthesis Flow Breaks Down (Tell Us the Story)

• Tell us about the most recent hard block that failed to close timing — what block was it and why was it critical? Options: CPU core, AI/DSP block, SerDes/PHY, Memory subsystem, Interconnect/NoC, Security/Encryption IP, Other (specify)
• How reproducible are the timing failures on that block (e.g., deterministic vs flaky runs)? Options: Deterministic every run, Mostly reproducible, Intermittent/flaky, Hard to reproduce
• What was the typical timing gap when the block failed (approximate worst-case slack or percent shortfall)? Options: <5% (small gap), 5–10%, 10–20%, >20%, Measured in ns (enter below)
• How do you usually discover the failure — CI regression, manual signoff, silicon bring-up, or another way? Options: Automated CI/regression, Designer signoff run, P&R signoff results, Silicon validation/bring-up, Customer/partner report, Other
• What immediate steps does your team take when that block fails timing (who acts, what tools, timelines)?
• Which stakeholder groups get pulled in when that failure happens and how does that change decision velocity? Options: Block owner only, Block owner + CAD, CAD + Physical design, CAD + Management + Physical, Executive escalation required

Tooling and Custom Glue: What's Holding the Pipeline Together?

• If you had to name your synthesis & analysis toolchain in order, what does it look like today? Options: Synopsys DC + In-house scripts, Cadence Genus + Mentor tools, Siemens EDA flow + custom tools, Mixed-vendor (hybrid), All in-house/academic tools, Other (list)
• Which of these integration points cause the most friction? Options: Constraint handoff, Netlist/format conversions, Timing model mismatches, Script/CI fragility, Library/LEF/DB sync, Tool version drift, Other
• What custom scripts, wrappers, or engineers’ hacks are part of your flow (and how critical are they)?
• Do you run synthesis inside an automated CI (nightly/regression) or as ad-hoc engineer runs? Options: Full automated CI for synthesis, Partial CI (selected blocks), Mostly ad-hoc/manual runs, No formal CI for synthesis yet
• How strict are your version and reproducibility controls (docker/containers, pinned tool versions, golden scripts)? Options: Strict and enforced, Moderate (some pinning), Ad-hoc and inconsistent, Non-existent

IP Hotspots: Which Blocks Keep You Up at Night?

• Which IP blocks would you identify as recurring pain points across projects? Options: CPU subsystem, AI/DSP accelerators, High-speed IO/SerDes, Memory controllers, NoC/interconnect, Mixed-signal interfaces, Other (specify)
• For those hotspots, how often do they require manual tuning to meet timing (percentage of projects)? Options: Every project, Most projects, Some projects, Rarely, Never
• What is the typical impact when one of these hotspots fails timing (schedule slip, area increase, increased power, re-spin risk)? Options: Schedule slip (weeks), Area growth, Power budget breach, Increased verification time, Silicon re-spin risk, Other
• How much institutional knowledge is required to tune those blocks (single expert vs. team practice)? Options: Single go-to expert, Small team with expertise, Distributed knowledge across teams, No one has stable expertise
• Share an example of a hotspot where you achieved a breakthrough — what changed and why did it work?

Assumptions That Might Be Costing You Weeks

• Which beliefs do you hold about synthesis that you’d defend in front of execs (e.g., 'PPA hit only solvable in P&R')?
• Which of these assumptions do you suspect are most likely false or at least incomplete? Options: Place-and-route surprises are unavoidable, Synthesis cannot predict congestion, ML-driven heuristics are risky, Migration requalification always outweighs QoR gains, Tools are interchangeable
• What evidence have you collected that supports or contradicts those assumptions (benchmarks, regressions, silicon results)?
• How open is leadership to a short, low-risk experiment that challenges one of those assumptions? Options: Very open, Somewhat open, Skeptical but possible, Not open
• If you were to test one assumption with minimal effort, which would it be and why?

Reality to Pilot: Small Tests That Answer Big Questions

• If a 2–3 block pilot could prove or disprove months of requalification effort, which blocks would you pick? Options: Most timing-critical block, Largest area/power consumer, Flakiest/reproducibility issue block, Representative mid-size block, Blocks with existing golden regressions, Other (specify)
• Do you have reproducible testbenches, RTL freezes, and PDK-ready libraries available to run a pilot? Options: All ready today, Mostly ready with minor cleanup, Some artifacts exist but need work, Significant preparation required
• What would success look like for a pilot (pick top 3 acceptance criteria)? Options: Meet or beat timing on hardest block, Reduce ECO iterations by X%, Faster turnaround/runtime, No new CDC/lint regressions, Smooth migration of scripts, Low requalification overhead
• What internal approvals or sign-offs are required to greenlight a pilot and who holds them?
• How much migration/time budget can you realistically allocate to a pilot (weeks of CAD effort)? Options: <1 week, 1–2 weeks, 3–4 weeks, 1–2 months, More than 2 months

Practical Constraints and Risk Tolerances

• Which risks worry you most when evaluating a new front-end tool? Options: Full-methodology requalification cost, Script/flow migration effort, Incompatibility with PDK/LEF/LIB, Loss of IP performance, CAD team ramp time, Support quality from vendor, Other
• How tolerant is your organization to short-term regression if long-term QoR improves? Options: High tolerance (long-term payoff preferred), Moderate tolerance, Low tolerance (must be safe), Zero tolerance
• What timeline would be acceptable to evaluate and decide on a new synthesis flow (select one)? Options: 2–4 weeks (very fast), 1–2 months, 3–6 months, Longer than 6 months
• Who needs to be convinced to accept a pilot outcome as sufficient evidence (roles/titles)?
• If a pilot shows a 10–20% improvement in iterations to closure, how likely are you to invest in broader migration? Options: Very likely, Somewhat likely, Uncertain, Unlikely
• What support would make migration tolerable (pick all that apply)? Options: Migration scripts and templates, On-site CAD support, Training for engineers, Automated conversion tools, Co-development partnership, Other

Final Step: What We’d Need to Move Forward

• Which single KPI would convince you this discovery was worth your time (pick one)? Options: Reduced ECO iterations, Improved WNS/TNS, Lower runtime/turnaround, Reduced manual CAD hours, Smaller area/power at same timing
• What is the best next step in your view (select the most realistic)? Options: Run a 2-block pilot, Share logs/benchmarks for analysis, Schedule a technical deep-dive, Proof-of-concept run on our hardware, Not ready to proceed
• Who should be the main point of contact to coordinate artifacts, schedules, and approvals (name, role, email preferred)?
• Any final constraints, non-negotiables, or red lines we should know before proposing a pilot plan?
• How soon could your team feasibly start a small pilot if we agreed on scope and NDAs? Options: Immediately (this week), Within 2 weeks, Within a month, 2+ months, Unsure

Quick Check-in: Who’s in the Room?

• Who will be the core sponsor and the day-to-day CAD owner for this pilot? (Name + role)
• Which stakeholders should we include on technical updates and escalation (pick all who apply)? Options: VP Engineering/CTO, CAD Manager, Lead RTL Designer, Place & Route Owner, Sign-off/Validation Engineer, Security/IT, Procurement/Legal
• What time zone(s) and weekly windows are best for synchronized pilot runs and live troubleshooting? Options: PST (US West), MST / CST (US), EST (US East), UTC / GMT, CET, IST, SGT / HKT, Other
• What’s your preferred channel for urgent pilot communications (e.g., Slack channel, Teams, email, pager)? Options: Slack/Workspace channel, Microsoft Teams, Email, Phone/SMS, Ticketing system (Jira/ServiceNow), Other
• How confident are you that the listed CAD owner has the authority to approve tooling access and data sharing for the pilot? Options: Fully authorized, Authorized but needs approvals, Not sure—requires escalation, No authority

If the Pilot Stumbles, Who Pulls the Emergency Cord?

• Imagine a blocker causes the pilot to miss a scheduled run — who is empowered to halt or pause the pilot and why would they do so?
• What escalation path do you currently use for toolchain outages or run failures? Options: Internal CAD escalation, IT/Infrastructure, Vendor support on-call, Executive escalation, No formal path
• What SLA or response time would you expect from our support team for high-severity pilot issues? Options: <1 hour, <4 hours, <1 business day, 48+ hours
• Tell us about the last pilot or tool migration that required emergency rollback—what caused it and how long to recover?
• What emotional impact (stress, schedule risk, reputational risk) do unresolved pilot incidents have on your team and leadership? Options: High (major schedule/reputation risk), Moderate (extra work, slide deadlines), Low (manageable), Minimal/none

Do Your Environments Actually Mirror Tapeout Reality?

• How closely does your pilot environment(s) reflect the final tapeout environment in terms of libraries, PDKs, and floorplan constraints? Options: Identical parity, High parity (minor differences), Partial parity (key libs only), Low parity / synthetic
• Which environment types will you make available for the pilot (select all that apply)? Options: Local CAD cluster, On-prem server farm, Private cloud, Public cloud (AWS/Azure/GCP), HPC facility, Emulated/smaller datasets only
• Are there licensing or toolchain dependencies that could prevent full-parity runs (third-party analysers, proprietary scripts)? Please list.
• Do you have physical design data we need (placement/power/CTS snapshots) available for back-annotation during synthesis validation? Options: Full snapshots available, Partial snapshots available, Can generate on request, Not available
• What constraints or size limits (max cells, memory, runtime) have historically forced you to trim runs or use smaller testcases?

Where Could Access or Data Silos Quietly Kill the Pilot?

• Who owns the IP, NDA, and permission gates for the blocks we’ll need, and how quickly can they approve data transfers? Options: CAD owner, IP owner (design lead), Legal/NDA team, Security/IT, Unknown
• How will you provide the design snapshots and test vectors: secure SCP, S3 bucket, physical media, VPN, or other method? Options: Secure SCP/SFTP, S3 (pre-signed URLs), Private VPN transfer, Physical encrypted drive, Other
• Are there any IP blocks or datasets that cannot leave your network or must be anonymized/sanitized before transfer? Options: Yes — cannot leave, Yes — must be sanitized, No restrictions, Unsure
• What internal approvals or security reviews typically add the most delay when provisioning access for external vendors? Options: Formal security review, NDA/legal sign-off, IT firewall/VPN provisioning, Procurement/contract, None
• Estimate how long it will take to provision the required data access and accounts if approvals proceed normally. Options: <2 business days, 3–7 business days, 1–3 weeks, Longer / high uncertainty

If We Run Your Hardest Block, What WILL Break?

• List the one to three IP blocks you consider the most timing-constrained or highest-risk for this pilot (names, size, why they matter).
• From past experience, which failure modes show up first during synthesis or early signoff (pick all that apply)? Options: Timing ECO loops, Clock-domain issues/CDC, Physical congestion post-synthesis, Power/IR violations, Scripting/flow errors, Tool hangs/crashes
• Do you already have gold-standard test vectors and regression suites for those blocks, or will we need to build/clean them together? Options: Fully available, Partial—need augmentation, Need to create from scratch, Unsure
• What is an acceptable turnaround for a full-run of a hardest block during the pilot (max wall-clock hours)? Options: <4 hours, 4–12 hours, 12–24 hours, 24–72 hours, >72 hours
• What specific PPA or QoR delta versus your current flow would you consider a pilot success for those blocks? Be quantitative if possible.

Who's Responsible for Rollback, Validation, and Sign‑Off?

• If a pilot run causes regressions, what rollback mechanisms are currently in place and who executes them? Options: Automated snapshot rollback, Manual restore by CAD, IT/Infra rollback, No formal rollback
• Which validation gates must be executed before leadership signs off on pilot results (select all that apply)? Options: Functional regressions passed, Timing within target, Lint/CDC clean, Power within budget, Runtime and resource metrics validated, Migration scripts vetted
• Who on your team approves moving from pilot to expanded validation or broader adoption, and what evidence do they require?
• How do you want failures and passes documented—automated dashboards, emailed reports, or formal review decks? Options: Automated dashboard, Email summary + logs, Formal review deck with exec summary, All of the above
• What acceptance thresholds (timing slack, area, power, runtime) are non-negotiable for sign-off?

What Would Make You Sleep Easier During the Pilot?

• If you could ask our team to guarantee one thing about pilot execution, what would it be?
• Which of these support models would best reduce your risk during the pilot? Options: Dedicated on-site/remote CAD enablement, On-call engineering support, Weekly sync + run reviews, Runbook + automated notifications, Training sessions only
• Would you want a dry-run or smoke test with synthetic data before any production block is run? If so, how thorough should it be? Options: Yes — brief smoke test (functional), Yes — full-parity dry-run, No, proceed with real runs, Unsure
• What’s the single biggest emotional concern leadership expresses about changing front-end tools (e.g., schedule slip, requalification effort, IP safety)? Options: Schedule slip, Requalification cost, Security/IP leakage, Loss of local expertise, Vendor lock-in risk, Other
• What early-warning metric would immediately reassure you the pilot is on track (e.g., first-run timing delta, runtime within x%, successful smoke test)?

Let's Schedule the First Run — Are We Missing Anything?

• What is your preferred target window to start the pilot (pick one)? Options: This week, Next week, In 2–4 weeks, In 1–3 months, TBD—dependent on approvals
• Before the first full run, which of these checklist items must be completed (select all that apply)? Options: NDA/contract signed, Data/PDK access provisioned, Licenses available, Run scripts validated, Stakeholder kick-off completed, Smoke test passed
• How long of a runway (in days) do you need from kickoff to first full-block result, realistically? Options: 1–3 days, 4–7 days, 8–14 days, 2–4 weeks, Longer
• Who should be notified immediately when the first run completes or fails (names/roles)?
• Are there any hard blackout dates (tapeout milestones, freeze windows) during which pilot runs cannot occur? Options: Yes — provide dates, No blackout dates, Unknown
• What deliverables do you need after the pilot’s first run to feel ready for a review (e.g., timing report, PPA comparison, logs, migration script snapshot)? Options: Timing report, PPA comparison, Runtime/resource report, Lint/CDC results, Raw logs + scripts, Executive one-page

Warm up — What's the single thing keeping you up at night?

• What's the primary reason you're exploring a new synthesis/verification flow right now? Options: Missed timing closure on a critical block, Excessive manual ECO cycles delaying tapeout, Competitor achieved PPA advantage, Runtime doesn't scale to full chip, Toolchain aging / vendor fit issues, Other
• Who on your team is most often asked to fix timing or lead synthesis escalations (title/role)? Options: VP Design Engineering, CAD Manager, Principal CAD Engineer, Block Owner / Lead Designer, EDA Tool Owner, Other
• How soon do you need meaningful improvement for your current project schedule? Options: Immediately (weeks), 1-3 months, 3-6 months, 6-12 months, Unsure
• In one short paragraph, describe a recent timing failure or ECO loop that cost schedule or resources (what happened, impact).
• Roughly how many CAD/design FTEs are tied up in synthesis-to-P&R iterations on your most-critical blocks? Options: 1-2, 3-5, 6-10, 11-20, 20+

What if your synthesis could read the physical map before layout?

• Where in your current flow do you most frequently discover physical congestion or hold issues—during synthesis, early P&R, or late STA? Options: During synthesis, Early P&R/place, Post-place STA, During signoff STA, Other/varies
• Walk me through your synthesis → P&R → STA loop for the hardest block: what are the concrete steps, tools, and handoffs?
• Which tools power each step in that loop (synthesis, placement, STA, lint/CDC, gate-level sim)? Select all that apply. Options: Our incumbent synthesis, Competitor synthesis, Open-source tools, Internal scripts/wrappers, Commercial P&R tool A, Commercial P&R tool B, Custom STA flow, Other
• How many ECO iterations does a typical critical-block fix require from first failing timing to signoff? Options: 1-2, 3-4, 5-7, 8-10, 10+
• When these loops occur, what emotions or organizational effects show up (urgency, blame, overtime, delayed launches)? Give a real example.

Are you quietly accepting a persistent PPA gap—and what would it take to stop that?

• What are your target QoR metrics for a successful synthesis: timing slack, area, and power targets (give numbers or ranges)?
• Which single metric would you prioritize if you had to trade between timing, area, power, and runtime? Options: Timing (QoR), Area, Power, Turnaround/runtime, Lint/CDC accuracy
• Do you already have benchmark blocks and golden runs we should match or beat? If so, how many and how representative are they? Options: Yes — 1-2 blocks, Yes — 3-5 blocks, Yes — >5 blocks, No benchmark blocks yet, In progress
• Share a recent benchmark result (timing PPA and runtime) and what you felt was most surprising about it.
• How tolerant are you to small regressions in one metric if others improve (e.g., 2% area increase for 10% timing gain)? Options: Very tolerant, Somewhat tolerant, Prefer neutral tradeoffs, Not tolerant

What would it actually cost to requalify a new front-end—not hypotheticals, the real bill?

• Estimate the internal effort to migrate flows and scripts if you switched synthesis engines (CAD FTE-months). Options: <1 FTE-month, 1-3 FTE-months, 4-6 FTE-months, 7-12 FTE-months, >12 FTE-months
• Which parts of migration feel the riskiest to you—scripts, PDK integration, VCS/GL simulation, signoff criteria, or cultural buy-in? Options: Script migration, PDK/license setup, Gate-level regression, Signoff flow changes, Team training, Other
• Tell us about a past tool migration that went poorly—what broke and what would you change if you could redo it?
• What contractual or commercial protections would make you comfortable trying a pilot (credits, rollback terms, support SLAs)? Options: Free pilot credits, Rollback clause, Defined success gates, Dedicated support engineer, Fixed-price migration package, Other
• How will your organization judge whether the requalification effort was worth it (who signs off and what documentation is required)?

Show us the worst block — and help us prioritize where success matters most

• Which block or IP do you consider the single hardest for synthesis right now (name or functional description)?
• For that block, supply key characteristics we need to reproduce the challenge: FF count, macro usage, number of clocks/domains, and presence of macros/leaf cells.
• Which process node and PDK does that block target? Options: 7nm, 5nm, 3nm, 14/16nm, 28nm/older, Multiple
• Are you able to share sample RTL, constraints, golden results, and run scripts for a pilot? (Select option to indicate access level.) Options: Yes — shareable under NDA, Yes — internal sandbox only, No — cannot share but can run tests internally, Need legal/exec approval
• If sharing is possible, what's the expected timeline for getting us the data and who will be the point of contact?

If this pilot delivered its promises next quarter, what would actually change?

• Imagine the pilot hits your success gates: how would those gains translate into schedule, headcount, or product competitiveness?
• Which downstream stakeholders would feel the biggest relief (e.g., tapeout program manager, SoC architect, firmware team)? Options: Program Manager, SoC Architect, Verification Lead, Firmware/Driver Team, Marketing/Business
• What are the non-technical success signals that matter—e.g., executive confidence, reduced fire-drills, or faster design handoffs?
• If the pilot shows mixed results (wins on timing but runtime is worse), where would you want the emphasis placed to decide next steps? Options: Optimize QoR first, Optimize runtime first, Balance both equally, Need further analysis
• How would you like success to be reported—raw PPA tables, side-by-side signoff runs, annotated runbooks, or recorded sessions? Options: Side-by-side PPA reports, Signoff scripts and logs, Annotated runbook with steps, Recorded walkthroughs/demos, Executive summary

Who needs convincing — and what will it take to win them over?

• List the decision-makers and influencers for a tool change and the key concern each would raise (name/role + concern).
• Which of these is the single biggest blocker to approval today: technical risk, headcount, cost, procurement complexity, or executive buy-in? Options: Technical risk, Headcount/effort, Cost, Procurement, Executive buy-in
• How does procurement typically structure enterprise EDA purchases (pilot → enterprise, PO timeline, legal reviews)? Options: Pilot then enterprise, Direct enterprise purchase, Evaluated by committee, Other/varies
• What commercial or support model would make your stakeholders comfortable—term license, perpetual, usage-based, or fixed milestone payments? Options: Term license, Perpetual, Usage-based, Fixed-milestone, Subscription
• What is the expected procurement timeline from pilot approval to a signed enterprise agreement? Options: <1 month, 1-3 months, 3-6 months, 6-12 months, Unsure

Let’s design a pilot you can’t say no to

• Which environments and access will we need to run a representative pilot: PDKs, licenses, cluster access, design database snapshots? Options: PDK access, Tool licenses, Compute cluster, Design DB snapshots, Internal CI system, Other
• Who will own day-to-day execution, and who will validate results on your side (names/roles)?
• Define the acceptance checklist you would insist on for a successful pilot (timing thresholds, runtime limits, lint/CDC pass criteria).
• What level of vendor support do you expect during the pilot—dedicated engineer, on-site help, weekly syncs, or written runbooks? Options: Dedicated engineer, Remote support, On-site assistance, Weekly syncs, Written runbooks
• What would be a reasonable pilot duration and cadence for checkpoint reviews? Options: 1 week, 2 weeks, 1 month, 2-3 months, Custom cadence
• If the pilot succeeds, what is your preferred next step: phased rollout, parallel validation, or immediate migration? Options: Phased rollout, Parallel validation, Immediate migration, Further pilot/scale tests

Final check — practical constraints and immediate next moves

• What legal, IP, or compliance constraints should we know about before requesting design assets?
• Do you have any compute or licensing limits that would prevent us from running the pilot at scale? If yes, describe. Options: No limits, Limited compute, License constraints, Both compute and license limits, Unsure
• Who should receive the pilot proposal and technical plan on your side (name, role, email)?
• If we proposed a pilot today with a proposed start date, how soon could you mobilize the necessary resources? Options: Within 1 week, 2-4 weeks, 1-2 months, Longer / need planning
• What would make you hesitate to proceed after reviewing a pilot plan, and how can we remove that hesitation upfront?