Industrial & Manufacturing
›
Aerospace & Space
›
Commercial Space
Remote Sensing
Zero-failure programs where certification, partners, and supply chains must execute against gated evidence.
Planet Labs
Maxar Technologies
BlackSky
Satellogic
Inside this journey
-
Pre-Discovery
Align stakeholders, decision roles, timelines, and procurement readiness before technical discovery.
-
Stakeholder Alignment
Confirm decision roles, timeline, procurement constraints, and required approvals across program, IT, and operations.
Alignment Questions
Starting Together: Quick Context
- To begin, what's the primary mission or use case you're evaluating imagery for right now?
- Which team or role is leading this evaluation inside your organization?
- Roughly how many Areas of Interest (AOIs) or discrete sites are you actively monitoring?
- What cadence do you currently need or expect across those AOIs?
- What's the single most urgent problem you'd like better imagery or analytics to solve for you right now?
Are You Settling for 'Good Enough' Imagery?
- What compromises are you currently accepting because 'perfect' imagery feels unattainable?
- How often do those compromises (lower resolution, longer revisit, cloud gaps, analytics uncertainty) lead to missed detections or wrong decisions?
- Which trade-offs are you most frequently forced to accept?
- Tell us about a recent incident where an imagery limitation had an operational consequence—what happened and what was the impact?
- How long have you been working within these constraints?
Where the Work Actually Breaks Down
- If your imagery pipeline stopped performing at the level you need tomorrow, where would the pain show up first?
- How do imagery delivery problems most commonly surface in your workflows?
- Which GIS or analytics platforms must imagery feed reliably into for your team to function?
- How is imagery ingested into your stack today?
- What percentage of imagery over your AOIs tends to be cloudy or otherwise unusable (your best estimate)?
- How quickly would repeated analytics inaccuracies erode trust with your internal stakeholders?
When Data Decides Your Next Move
- Imagine imagery could tell you with confidence when to act—are your operational processes organized to move on that signal?
- Which measurable signals make imagery 'actionable' for you?
- What is the maximum acceptable latency from collection to delivery for an operational decision?
- What minimum spatial resolution do you require to meet your objectives?
- How do you currently validate imagery and analytics accuracy (ground truth, field checks, third-party datasets)?
- How will you quantify mission impact or ROI from improved imagery and analytics?
What Would Winning Look Like?
- If this engagement were labeled a success in your next program review, what headline or outcome would you want to see?
- List the top three outcomes that would make the project an undeniable win for your team.
- Which stakeholders must be satisfied before you can move from pilot to production?
- What concrete acceptance criteria would you require for image quality, analytics accuracy, and delivery timelines?
- Are there legal, sovereignty, or classification constraints we must design around?
- What pilot timeline would let you confidently evaluate image quality, revisit, and analytics (select preferred duration)?
Barriers Between You and Reliable Answers
- What single organizational blocker is most likely to kill this project faster than any technical problem?
- How long does procurement and contracting typically take for new data or pilot purchases?
- Which security or compliance certifications do you require vendors to meet?
- What is your budget posture for this work (pilot already funded, need to request pilot funds, enterprise budget available, TBD)?
- Who are the decision-makers and what criteria will they weight most heavily when deciding to proceed?
- What would most convince your IT/security teams that a vendor is low risk (evidence, architectures, references)?
Practical Steps to Prove It
- If we could reduce risk to near-zero for the first 60 days, what would you want us to demonstrate?
- Which pilot structure would best prove value to your stakeholders?
- Which AOIs, seasons, or scenarios should we include in the pilot to validate end-to-end value?
- What specific success metrics and thresholds would you require at pilot close to green-light production?
- Which technical prerequisites must be in place before we start (API keys, cloud bucket access, IP whitelisting, sample ingestion)?
- How would you prefer progress to be reported during the pilot?
- Realistically, when could you start a pilot if approvals were completed?
-
Current State Mapping
Document existing data sources, GIS workflows, pain points, AOIs, and operational constraints that affect imagery use.
Current State
Walk Me Through a Day in Your Map Room
- What’s the single most frequent task your team performs with geospatial data today?
- Which GIS platforms, desktop tools, and analysis environments do you rely on most?
- List the primary sources you ingest (satellite constellations, drone imagery, aerial, ground sensors, government rasters, third‑party feeds).
- How do you currently manage and track Areas of Interest (AOIs)?
- Roughly how many active AOIs or monitoring polygons does your team maintain right now?
What Are We Missing That Keeps You Guessing?
- What recurring blind spot have you tolerated because you haven’t found a reliable imagery solution?
- Which kinds of change or events most often go undetected with your current imagery and analytics?
- How often do missed detections or poor image quality lead to operational rework, false alarms, or missed decision windows?
- Tell us about a recent incident where imagery limitations had real consequences—what happened and what was at stake?
- How do you currently confirm whether a flagged change was real (ground truth, partner intel, follow-up tasking)?
Where Does Your Pipeline Snap?
- At what step in your ingestion → analysis → delivery pipeline do imagery or analytics most often fail to be useful?
- Which file formats and delivery mechanisms cause the most friction for your engineers or GIS team?
- Do you have automated ingestion pipelines? If yes, which steps remain manual (naming, reprojection, cloud masking, metadata tagging)?
- What is your acceptable end‑to‑end latency from tasking to GIS‑ready asset for urgent workflows?
- How often do you require pre‑processed imagery (orthorectified, radiometrically corrected, cloud masked) versus raw imagery?
Which Places Keep You Up at Night?
- If visibility disappeared over one of your AOIs for a week, which AOI would cause the most operational risk—and why?
- Which AOI categories are top priority for you right now?
- For your highest‑priority AOIs, what is the smallest object or change you must reliably detect (in meters)?
- What revisit cadence do your top AOIs require to be operationally useful?
- Which environmental conditions most frequently reduce imagery usefulness for these AOIs?
Who Needs to See This — and Who Holds the Keys?
- Who in your organization would stop this purchase if they weren’t convinced—who holds veto power or budget control?
- Which internal teams will actively use the imagery and analytics outputs?
- What approvals, security reviews, or clearances are needed before we can deliver imagery to your environment?
- Which delivery patterns align with your security and operations needs?
What Would Make You Sleep Easier About Data Quality?
- What specific image or analytics failure would cause you to reject a dataset without further review?
- What is the minimum spatial resolution required for routine decision making?
- How do you quantify acceptable cloud cover for an image to be considered usable?
- Which analytics outputs are must‑have vs. nice‑to‑have for your workflows?
- How will you measure success in a pilot (specific KPIs, e.g., detection precision/recall, latency, integration time)?
If We Dropped a Test Dataset on Your Desk Tomorrow…
- Would a one‑off sample meaningfully resolve your main technical doubts, or would it prompt additional questions? Explain which outcomes you’d need.
- Which sample deliverables and formats would you need to run a real evaluation?
- How long would your team need to validate imagery quality and analytics against ground truth or reference data?
- Who on your team would own the evaluation, and who signs off on pass/fail decisions?
- What integration test would convince your engineers our delivery can plug into your architecture (specific endpoints, auth patterns, ingestion scripts)?
What Could Stop This Project Cold?
- What single operational, legal, or procurement issue is most likely to pause or cancel this initiative in the first 90 days?
- Are there regulatory, export control, or data residency requirements we must design around?
- Does your IT or security team restrict certain endpoints, protocols, or cloud providers we should know about?
- What limitations in bandwidth, storage, or on‑prem resources would impact frequent delivery of large imagery files?
What Would a Small Win Look Like?
- If a 30–60 day pilot succeeded, what concrete outcomes would make you consider scaling to a subscription?
- Which pilot KPIs would most influence your decision (pick up to three)?
- Which commercial model best fits how you’ll budget this work?
- Realistically, what timeline would your organization need to decide to move from pilot to program?
- Who are the essential stakeholders we should include in pilot planning and scope definition?
-
-
Outcome Discovery
Define target outcomes, measurable success signals (revisit, resolution, latency), and acceptance criteria for imagery and analytics.
Discovery Questions
Start With the Outcome You’d Celebrate
- What one concrete outcome would make you say this imagery program was an unequivocal success?
- Which stakeholders would celebrate that outcome and why (program, IT, ops, external regulators)?
- How soon would you expect to see that outcome after starting a pilot (realistic timeline)?
- If we delivered that outcome, what would the immediate operational change look like (what a person on the team would do differently)?
- Who has final acceptance authority for that outcome, and what format of evidence do they require (report, dashboard, live demo)?
Why Most 'Success Metrics' Miss the Point
- Which commonly quoted metric (revisit, resolution, accuracy, latency, cost) do you suspect gives a false sense of readiness for your use case?
- Can you share a specific example where a vendor’s metric looked good on paper but didn’t translate to usable results in the field?
- How do you currently discover those gaps—through manual QA, end-user complaints, missed detections, or some other channel?
- When those gaps appear, what are the real operational consequences (missed actions, delayed decisions, false alarms, wasted labor)?
- How comfortable are you with a vendor that admits limitations up front and designs around them?
The Hard Numbers We Need to Win
- If you had to pick the three KPIs we must prove, which would they be?
- For revisit cadence, what are the acceptable ranges for your priority AOIs?
- For spatial resolution, what is the minimum ground sample distance (GSD) that lets you make required detections or measurements?
- What is the maximum end-to-end delivery latency (tasking to usable file/analytics) that still supports timely decisions?
- How do you want analytics accuracy expressed and validated (precision/recall, confusion matrix, sample-based error rate)?
- Please list numeric thresholds (for the KPIs above) that would make you feel confident—use AOI-specific values if needed.
What Fails Quietly—Acceptance Criteria That Hide Risk
- Where have pilots or QA checks passed but production use later revealed systemic issues?
- Which failure modes worry you most when accepting imagery/analytics (e.g., seasonal confusers, edge cases, cloud bias, drift over time)?
- What minimum sample size, diversity of conditions (season, time of day, weather), and AOI distribution do you require to be confident acceptance isn’t luck?
- How much geolocation error (meters) is tolerable before imagery becomes operationally unusable for you?
- What acceptance evidence do you need beyond numbers—annotated examples, side-by-side comparisons, or field-verified samples?
The Imagery & Analytics Experience You’d Trust
- Think of a time an image or analytic made you act immediately—what qualities of that output gave you confidence?
- Which delivery formats and interfaces do your systems and analysts require to use data without heavy rework?
- What ingestion or preprocessing constraints should we plan for (coordinate systems, metadata schemas, tile sizes)?
- How important is visual evidence (native imagery) versus derived layers (classified objects, change masks) for your decision process?
- What turnaround for a validated sample (from delivery to analyst confirmation) feels reasonable during a pilot?
Operational Constraints and Non‑Negotiables
- What non-negotiable operational constraint would cause you to halt the engagement immediately?
- Do you have procurement, security, or legal timelines that must be met before data can be used in production?
- What minimum security/compliance standards must we meet (e.g., FedRAMP, ITAR, ISO 27001, on-prem keys)?
- Are there internal data rights or export restrictions that would limit sharing imagery or analytics with third parties?
- How would procurement milestones affect pilot scope and timing (e.g., PO in place before delivery)?
Pilot, Acceptance, and Handoff—How We Prove It
- What three acceptance tests must pass during a 30–60 day pilot for you to approve moving to subscription?
- Which AOIs or scenarios should we prioritize in the pilot to surface meaningful, representative results?
- Who will be responsible on your side for running acceptance tests and signing off (role names, not people)?
- What specific handoff artifacts do you expect at acceptance (runbook, API keys, sample datasets, model weights, integration scripts)?
- If acceptance uncovers issues, what remediation timeline would you require before altering the commercial commitment?
A Little About Feelings and Trust
- How does uncertainty about imagery or model quality affect your team’s willingness to act on the outputs?
- Have you had a prior vendor experience that damaged trust? What happened and how did it make your team feel?
- What vendor behaviors rebuild confidence fastest after a mistake (transparency, rapid fix, root-cause analysis, compensation)?
- How frequently would you like progress updates during the pilot to feel comfortable (weekly, biweekly, daily standups)?
- What level of co‑working or embedded support would make integration and acceptance feel low-risk (dedicated engineer, joint war room, documentation only)?
Next Steps, Owners, and Signals to Watch
- If we agreed to a 'no-surprises' pilot, what early warning signs in the first two weeks would make you nervous?
- Who should be our day-to-day point of contact and what’s their preferred escalation path?
- What immediate data or demo would you want to see from us to feel confident in proceeding to pilot planning?
- What internal milestone or meeting will be the decision point after the pilot (steering committee, ops review, procurement approval)?
- Is there anything else—hidden constraints, political dynamics, or desired outcomes—we should know now to avoid surprises later?
-
Solution Experience
Use the customer’s AOIs and scenarios to validate how imagery, analytics, and delivery timelines produce the required answers.
Experience Meetings
- Experience Prep: Current State, Consequence & Success Signals
- Sample Delivery & Initial Proof Using Customer AOIs
- End-to-End Delivery & Integration Test
- Analytics Accuracy, Edge Cases & Tuning Session
- Acceptance Review & Pilot Commitment Decision
- Obtain explicit customer sign-off on per-AOI acceptance criteria to enable the pilot.
- Annotate and record example 'proof slides' tying each analytic output to the customer problem statement.
- Review Delivery Architecture & Responsibilities
- Prove delivery latency and end-to-end ingestion meet the customer's acceptance criteria.
- Confirm integration responsibilities, API access, and any needed sandbox or production credentials.
- Agree on operational escalation and retry procedures for missed or delayed collects.
- Seller to provide test API keys, sample endpoints, and a data package for customer sandbox ingestion.
- Customer to run ingestion tests in their environment and report any format/metadata mismatches.
- Define SLAs and notification procedures for the pilot (delivery latency, uptime, incident response).
- Review Scenario-wise Accuracy Metrics
- Demonstrate analytics meet, or outline a concrete plan to meet, the customer's accuracy thresholds for each key scenario.
- Document failure modes and agree a prioritized tuning plan with timelines and owners.
- Introductions & Objectives
- Seller to run prioritized tuning experiments and deliver revised analytics for validation within agreed timeframe.
- Customer to provide additional labelled examples for the most problematic edge cases.
- Produce a per-AOI accuracy report that maps metrics to business consequence for pilot SOW.
- Summary: Findings vs Success Signals
- Achieve a mutual decision to proceed to a pilot with a signed SOW or agree a clear remediation plan if not ready.
- Assign owners, timelines, and acceptance test definitions for the pilot so execution can begin without ambiguity.
- Ensure all commercial/technical contingencies that would block the pilot are identified and addressed.
- Seller to draft pilot SOW and acceptance test checklist and circulate for customer review within agreed SLA.
- Customer to confirm budgetary approval and assign program/IT owners required for pilot kickoff.
- Schedule pilot kickoff meeting and technical onboarding within two weeks of SOW agreement.
- Surface and document a single-sentence current state and an explicit one-sentence future state tied to measurable success signals.
- Make the consequence of the current state explicit in operational/financial/risk terms so the experience is urgent.
- Agree on AOIs, scenarios, ground truth requirements, and the concrete pre-work and delivery schedule for sample data.
- Customer to deliver AOI shapefiles, scenario descriptions, and available ground-truth samples (labels) by agreed date.
- Seller to publish a concise success-signal checklist (revisit, resolution, latency, accuracy thresholds) for validation.
- Schedule hands-on validation sessions and assign owner for each AOI/scenario.
- Recap Success Signals & Acceptance Criteria
- Prove the offering produces outputs that map directly to the customer's future state for priority AOIs.
- Force explicit customer validation (yes/no with rationale) for each AOI/scenario reviewed.
- Identify and document gaps or tuning needed to meet acceptance criteria.
- Seller to deliver additional sample variants (different sensor, time-of-day, band combinations) for identified gaps.
- Customer to provide any missing ground-truth labels or schedule site visits for verification where required.
- Open Issues & Risk Register
- Walk Through Representative True/False Positives
- Delivery Overview (products & formats)
- One-sentence Current State
- Live/Recorded Request-to-Delivery Playback
- Edge Cases & Environmental Constraints
- Explicit Consequence
- GIS Ingestion Test
- Guided Walkthrough: AOI #1
- Pilot Scope Recommendation
- Define Future State & Success Signals
- Failover, Retry & Notification Behaviors
- Tuning & Customization Plan
- Decision & Commitment
- Guided Walkthrough: AOI #2 (if applicable)
- Decision Checkpoint
- Confirm AOIs, Scenarios & Validation Data
- Forced Validation & Customer Confirmation
- Validation Sign-off Criteria
- Assign Owners & Next Steps
- Capture Gaps & Next Adjustments
- Logistics & Pre-work
-
Solution Scope
Define collections (archive vs tasking), cadence, resolution, analytics modules, delivery formats, and integration responsibilities.
Scope Configuration
- Deliver Orthorectified Optical Archive Imagery
- Task Satellite and Deliver New Collection Over AOI
- Deliver Daily/Sub-daily Monitoring Imagery Feed
- Deliver Synthetic Aperture Radar (SAR) Imagery
- Deliver Automated Change Detection Layers
- Deliver Object Detection Annotations (Vessels/Vehicles)
- Deliver Vegetation Index (NDVI) Time-Series
- Deliver Time-Series Anomaly Detection Layer
- Deliver Cloud-Optimized GeoTIFFs via API
- Deliver NITF-Formatted Imagery with Full Metadata
- Deliver GIS Connectors (WMS/ArcGIS REST)
- Deliver API Documentation and Code Samples
- Deliver Real-Time Change Alert Webhook Feed
- Deliver Expedited Priority Tasking-to-Delivery
Scope Questions
Deliver Orthorectified Optical Archive Imagery
- Do you require imagery from a specific date range or historical window?
- If yes, specify earliest and latest acceptable dates (YYYY-MM-DD to YYYY-MM-DD) or describe the time window.
- What minimum ground sample distance (spatial resolution) is acceptable for archive imagery?
- What is the maximum allowable geolocation / orthorectification error (RMSE) for delivered imagery?
- Which delivery formats do you need for archive imagery?
Task Satellite and Deliver New Collection Over AOI
- Do you want one-time tasking or recurring tasking over the AOI?
- Provide AOI geometry and approximate area (attach file or describe extent: e.g., single site, 10 km2, country-scale).
- What target revisit cadence do you need for tasking (per AOI)?
- What sensor and quality constraints must be met for tasking (cloud cover %, sun angle, off-nadir limit)?
- Do you require guaranteed tasking SLAs (e.g., tasking within X hours, delivery within Y hours)? If yes, specify targets.
Deliver Daily/Sub-daily Monitoring Imagery Feed
- What is the geographic scope of the monitoring feed (points, corridors, polygons, country, global)?
- Desired revisit frequency for monitoring (select best fit)
- Which data products do you need in the feed (raw imagery, analytics overlays, metadata)?
- What maximum delivery latency is acceptable from capture to availability (minutes/hours)?
- How will you ingest the feed into your systems (API pull, cloud bucket push, S3/GS/NAS, FTP, other)?
Deliver Synthetic Aperture Radar (SAR) Imagery
- Do you need SAR for day/night or all-weather monitoring specifically?
- Which SAR modes/polarizations and resolutions do you require (e.g., single-pol, dual-pol, quad-pol; fine/coarse)?
- Do you require radiometrically calibrated products, terrain correction, and/or interferometric-ready products?
- What output formats and preprocessing levels do you expect (GeoTIFF, GRD, SLC, backscatter coefficients)?
- Are there specific incidence angle, acquisition geometry, or temporal baselines required for your use case?
Deliver Automated Change Detection Layers
- What types of changes must be detected (construction, vegetation loss, flooding, new vehicles, other)?
- What minimum detectable change threshold do you need (area in m², percent reflectance, object count)?
- Preferred output for change layers (binary mask, graded confidence map, vectorized polygons, time-stamped events)?
- How often should change detection run and be delivered (real-time/near-real-time, daily, weekly)?
- Do you require human analyst validation (hybrid) for detected changes or fully automated delivery?
Deliver Object Detection Annotations (Vessels/Vehicles)
- Which object classes are required (e.g., vessel types, vehicle types, containers)?
- What detection performance targets do you need (precision, recall, minimum IoU)?
- Do you need bounding boxes, segmentation masks, keypoints, or attribute classification (e.g., vessel type, flag)?
- What temporal/contextual constraints aid detection (time of day, sea state, seasonal patterns)?
- How should detection outputs be delivered (GeoJSON with confidence, annotated images, CSV manifests, API endpoint)?
Deliver Vegetation Index (NDVI) Time-Series
- What temporal resolution and historic depth do you require for NDVI time-series?
- Do you need cloud/shadow masking and gap-filling in the time series?
- Preferred delivery format for time-series (CSV/Parquet timeseries, cloud-optimized raster stacks, API endpoints)?
- Do you require per-pixel time-series or aggregated metrics per polygon (e.g., field-mean NDVI)?
- Are derived indices beyond NDVI required (EVI, SAVI, NDWI)? If so, list.
Deliver Time-Series Anomaly Detection Layer
- What anomaly types are relevant (vegetation stress, sudden area change, infrastructure degradation, unusual movement)?
- What baseline period should be used to define normal behavior (e.g., past 12 months, multi-year seasonal baseline)?
- Preferred output format for anomalies (GeoTIFF mask, vector alerts, ranked list with severity scores)?
- What false-positive tolerance or minimum anomaly magnitude is acceptable?
- Do you require analyst review or a feedback loop to retrain anomaly models?
Deliver Cloud-Optimized GeoTIFFs via API
- Which storage/access pattern do you prefer for COGs (S3 bucket, signed URLs, direct API streaming)?
- Do you need tile pyramids, overviews, and internal tiling optimized for web/GIS clients?
- What metadata and tags must be embedded in the COG (acquisition time, provenance, quality flags)?
- Do you require presigned/persistent URLs and access controls (IAM, token-based)?
- Are there bandwidth or regional residency constraints for hosting the COGs?
Deliver NITF-Formatted Imagery with Full Metadata
- Is NITF required for all deliveries or only for specific customers/exports?
- What mandatory metadata elements or extensions must be included in NITF (eg. STANAG fields, custom tags)?
- Do you require encryption, signing, or adherence to specific NITF security profiles?
- What downstream systems must be compatible with NITF outputs (catalog IDs, ingest scripts)?
- Do you require sample NITF files for validation prior to full delivery?
Deliver GIS Connectors (WMS/ArcGIS REST)
- Which connector types are required for your environment (WMS, WMTS, ArcGIS REST, WCS)?
- Do you require authenticated connectors (token/API key/OAuth) or public endpoints?
- What layer styles and coordinate reference systems (CRS) must be supported?
- Are there performance requirements (concurrent users, tile response time) for the connectors?
- Do you need examples or connector templates for direct import into ArcGIS/QGIS?
Deliver API Documentation and Code Samples
- Which programming languages or platforms should code samples cover (Python, JavaScript, R, C#)?
- Do you need step-by-step guides for authentication, sample queries, and ingest workflows?
- Should documentation include sample data packages and end-to-end integration examples for common GIS platforms?
- Do you require interactive API consoles (Swagger) or downloadable SDKs?
- Are there internal developer teams that need private docs or an onboarding sandbox environment?
-
Mutual Commit
Finalize commercial terms, SLAs, data rights, security requirements, and pilot vs subscription commitments.
Agreement Modules
- Non-Disclosure Agreement (NDA)
- Master Services Agreement (MSA)
- Statement of Work (SOW)
- Service Level Agreement (SLA)
- Pricing & Commercial Terms
- Data Licensing & Rights
- Security & Compliance Addendum
- Export Controls & Regulatory Compliance
- Pilot Agreement & Success Criteria
- Acceptance Test Plan
- Integration & Implementation Plan
- Payment Terms & Invoicing
- Change Order & Scope Amendment
- Termination, Exit & Data Return Plan
- Liability, Indemnity & Insurance
-
Deployment
Operationalize rollout with readiness checks, enablement, and outcome validation.
-
Pre-Deployment Readiness
Verify API keys, access controls, environments, sample data delivery, and integration prerequisites are in place.
Readiness Questions
Quick Intro: Who's on This Mission?
- Please share the primary program or team that will use the imagery and analytics.
- Which operational role will be the day-to-day owner of the feed or pilot?
- What are the top three geographic areas of interest (AOIs) we should consider for initial testing? Please list site names, coordinates, or bounding boxes.
- Do you currently have an active imagery subscription or archive access with any provider?
- How will you judge whether a pilot is successful (select up to three primary success signals)?
- Who should be our single technical point of contact for API keys, network changes, and test validations? Include name, role, and best contact info.
If Your Data Stopped, What Would Break First?
- If imagery or analytics stopped arriving on schedule for a week, what operational impact would you see?
- Which downstream systems would be impacted first (e.g., automated alerts, ML models, dashboards, field ops)?
- When a feed underperforms, how quickly does your team typically detect and escalate the issue?
- Can you describe a recent incident where imagery timing or quality prevented you from answering a mission question? What happened?
- How much latency or data loss can your mission tolerate before senior stakeholders demand emergency action?
What’s Hidden Behind Your Firewall?
- Which security controls would stop us from simply standing up a test connection today?
- Do you prefer data pushed into your environment (push) or pulled from our API (pull)?
- Which authentication methods does your team require for production APIs?
- Are there formal security or compliance approvals we must obtain (e.g., ATO, SOC review, export controls) before connecting?
- If we need to run a lightweight demo inside your environment, what sandbox or test environment would we use and who governs access?
Can You Stand Up a Connection in Days—or Will It Take Quarters?
- What is the shortest realistic timeline you could accept for connecting a pilot feed and validating data end-to-end?
- Which internal processes usually cause the longest delays (select all that apply)?
- Who must sign off on the pilot (technical lead, security officer, legal, program director)? Please list roles and expected review lead-times.
- Are there blackout periods or seasonal operations when integrations cannot be performed (e.g., harvest, surge ops)? If yes, when?
- When approvals typically slip, what has helped accelerate timelines in the past (pre-approved language, SOC reports, pilot contracts)?
How Do You Judge 'Good Enough' for Imagery and Feeds?
- If you had to pick the single metric that makes imagery acceptable, which would it be?
- Please specify minimum acceptable values for key metrics: spatial resolution (m), max delivery latency (hours), minimum cloud-free (%) and desired revisit cadence.
- Which delivery formats and transfer methods must we support for seamless ingestion into your stack?
- Which analytics outputs are mission-critical for the pilot (select all that apply)?
- How will you validate analytics accuracy during the pilot—automated thresholds, human review, or ground-truth comparisons?
- Would you accept a small sample delivery (2–5 scenes) for quick technical validation before a wider pilot?
Who Will Own This Internally—Really?
- If integration breaks at 2AM, who is expected to wake up and fix it?
- Which teams will require training to operate and interpret imagery and analytics?
- What SLAs and escalation timelines do you expect for pilot issues (response time, resolution time)?
- Who is expected to own long-term data retention and storage costs once the feed is productionized?
- How do you prefer to track issues and enhancement requests during the pilot (ticketing system, shared doc, Slack/Teams)?
- What internal change management steps will be required to transition from pilot to production?
Let’s Make the Integration Work: Practical Next Steps
- If we handed you a working API key and sample payload today, what is the first thing you'd try?
- Do you want sample data delivered via our staging endpoint or directly into your cloud storage (S3/GCS) for testing?
- Which of these prerequisites are already in place for integration? (select all that apply)
- Do you require NDAs, data-use agreements, or export-control paperwork before we share samples?
- What time window works best for a technical handoff and first integration call?
- Who will provide ground-truth labels or test annotations if we need them to validate analytics?
Final Readiness Check — Are We Ready To Launch a Pilot?
- Given everything we've discussed, what would stop you from launching a pilot this quarter?
- Rate your confidence that essential prerequisites (API keys, access, environment, sample delivery) can be met in your desired timeline.
- What is the earliest realistic start date for a 30–60 day pilot?
- Please list any final open questions, top risks, or blockers we should resolve before kickoff.
- Would you like a one-page technical readiness checklist sent to your POC to help accelerate approvals?
-
Deployment Enablement
Coordinate onboarding tasks, schedule integrations, assign owners, and execute the pilot or feed activation.
-
Validation Checklist
Execute acceptance tests for image quality, revisit cadence, analytics accuracy, delivery latency, and GIS ingestion.
Validation Questions
Why this imagery effort matters to you today
- What is the primary mission or program you are trying to achieve with new imagery or analytics right now?
- Who on your team will be most impacted day-to-day by the imagery and analytics we provide?
- Tell us a recent example where imagery or analytics changed (or could have changed) the outcome of an operation — what happened and what was at stake?
- How urgent is this need on a scale from immediate (within days) to exploratory (no fixed timeline)?
- Which of these best describes what success would feel like to you at the end of a 30–60 day pilot?
Who’s actually in the room — and who’s whispering offstage?
- If your procurement process were a pipeline, where do most deals get stuck today and why?
- List the decision roles and approvers we should expect to engage (names, titles, and role in decision).
- Which stakeholders must sign off on data security, export controls, or classification before you can ingest imagery?
- What is your typical procurement timeline from proof-of-concept to contract award?
- How does procurement budgeting work for imagery—single-line item, shared across tools, or funded by program holds?
Are you quietly tolerating gaps that put operations at risk?
- How often does the imagery cadence you currently have miss an event or change you care about?
- When imagery or analytics produce incorrect or missing results, how does that show up operationally (e.g., false alarms, missed detections, rework)?
- How much cloud cover or weather-related loss do you typically see across your AOIs and how disruptive is that?
- How confident are you in your current analytics accuracy against your ground truth or field reports?
- Describe a recent incident where imagery limitations directly affected timelines, costs, or safety. How did it feel to be in that situation?
- Which of these trade-offs have you accepted—often silently—because no one offered a better option?
Where does your data live — and why do integrations usually break hearts?
- If your team had to hand us a map of your current GIS/data architecture, where would the biggest friction points be?
- Which imagery formats and delivery methods must we support for a successful proof-of-concept?
- How does your team currently automate ingestion and QA—pipelines, scripts, or human ops—and who owns that work?
- Tell us about a time an integration went smoothly. What made it easy? Conversely, what usually trips teams up?
- What constraints must we design around—firewalls, restricted environments, STIGs, FIPS, or other security baselines?
- Which internal teams will need API keys, sample data, or access for testing during the pilot?
If we could deliver exactly what you wished for, what would it tell you?
- What are the specific, measurable success signals you will use to judge the pilot (examples: revisit rate, detection precision/recall, latency thresholds)?
- For each success signal you selected, what numeric threshold would you consider a pass?
- Which AOIs or scenario types are highest priority for proving value, and why those?
- How will improved imagery change decisions — faster alerts, fewer field dispatches, regulatory reporting — and who benefits most?
- What concerns would make you hesitate to declare the pilot a success even if numbers look good on paper?
- If outcomes match expectations, what does a scaled subscription or operational handoff look like to you?
How will you test whether the images and analytics really work for you?
- Which acceptance tests must pass during the pilot for you to proceed (choose all that apply)?
- For image quality evaluation, which artifacts or examples do you want us to provide (raw scenes, pan-sharpened, multispectral indices, sample analytics overlays)?
- How will you measure analytics accuracy—what ground truth, labeled datasets, or field checks can you provide?
- What's your acceptable delivery latency from tasking or capture to ingestion in your systems?
- Who will own the validation checklist on your side and how will we coordinate retesting if thresholds aren’t met?
What would make you confident enough to say yes?
- If you had to sign off on one minimal commercial commitment to begin (e.g., short pilot PO, data credits), which would feel reasonable?
- What contractual or legal must-haves will block a deal if they are missing (data rights, retention limits, export controls, indemnity)?
- How do you prefer to structure a pilot’s success review—single stakeholder demo, formal validation report, or joint operational exercise?
- What internal risks (budget, competing priorities, leadership changes) could derail a deal even if the pilot succeeds, and how long have those been present?
- Realistically, what is the next step you'd like from us after this discovery (technical deep-dive, sample delivery, proposal, stakeholder workshop)?
-
-
Success
Review outcomes against success signals, capture lessons learned, and maintain a shared issues and enhancement log.
Success Reviews
- Success Review & Metrics Validation
- Lessons Learned & Continuous Improvement
- Issues Triage & Enhancement Backlog Workshop
- Executive Outcomes & Renewal Recommendation
Issues & Enhancements
- Opening & Objectives
- Identify gaps with root causes and assign remediation owners and deadlines.
- Get stakeholder sign-off on AOIs that meet acceptance criteria.
- Produce a consolidated outcome report containing metrics, artifacts, AOI pass/fail status, and signed acceptance lines.
- List AOIs requiring remediation with assigned owners, root-cause notes, and target completion dates.
- Deliver supporting evidence package (sample GeoTIFFs, analytics outputs, API logs) to customer repository for audit.
- Opening & Framing (experience rules)
- Produce a concise list of evidence-backed lessons and their quantified consequences.
- Convert lessons into a prioritized improvement backlog with owners and timelines.
- Agree documentation and knowledge-transfer actions to prevent recurrence.
- Document the top 8 lessons learned with supporting evidence and consequence statements.
- Create a prioritized improvement backlog with impact/effort scores and assigned owners.
- Schedule necessary training sessions and update onboarding/integration guides based on lessons.
- Pre-check & Inventory Confirmation
- Classify and prioritize all open issues with clear severity and operational consequence.
- Define remediation approach (workaround vs fix), acceptance criteria, and schedule for each prioritized item.
- Assign owners and commit to communication and delivery timelines.
- Update the shared issue tracker with severity, acceptance tests, owner, and target remediation date for each item.
- Create sprint tickets for high-priority fixes and schedule engineering resources.
- Publish a customer-facing status note for any items that impact operational use and include planned mitigation.
- Executive Summary & Current State (one-line)
- Provide an executive-level, evidence-backed summary of program outcomes and ROI.
- Obtain a clear decision or agreed next steps on renewal, scaling, or contract modification.
- Align commercial and operational owners on actions required to implement the decision.
- Prepare and distribute an executive brief with ROI calculations, outcome summary, and recommendation.
- If approved, draft contract amendment or renewal terms and circulate to procurement/legal for review.
- Schedule kickoff for scaled deployment or transition plan with assigned owners and a 90-day milestone plan.
- Validate measured outcomes against each success signal and produce a clear pass/fail for every AOI.
- Capture evidence (metrics + artifacts) that proves the future state where acceptance is achieved.
- What Worked — Evidence-based
- Reproduce & Evidence Review
- Current State (one-sentence)
- Business Consequence & ROI Assessment
- Outcomes vs Success Signals (high-level)
- Consequence Recap (one-sentence)
- Severity, Impact, and Consequence Assessment
- What Didn't Work — Consequences & Examples
- Customer Validation & Testimonials
- Remediation Options and Workarounds
- Future State Reminder (one-sentence)
- Root Cause Themes
- Metrics Presentation
- Prioritization & Sprint Planning
- Improvement Opportunities & Solutions
- Recommendation & Options (renew/scale/modify)
- Evidence Walkthrough (sample artifacts)
- Commercial & Contract Considerations
- Prioritization & Roadmap Inputs
- Define Acceptance Criteria & Test Cases
- Gap Analysis & Root Causes
- Documentation & Knowledge Transfer Plan
- Commitments & Communications Plan
- Decision & Next Steps
- Validation & Agreement per AOI
- Next Steps and Ownership