Answer
Prevent designers from chasing missing site information by making the site assessor responsible for a defined evidence package, not just a completed form. Use stage-specific acceptance rules, identifiable photos and measurements, explicit evidence states, a receiving decision, owned exception tickets, and change control. Designers should return, limit, or escalate work instead of privately reconstructing the survey.
Illustrative submission, not a customer case or measured result: A site assessor uploads a set of photos, marks the survey complete, and moves to the next appointment. The designer opens the package and still cannot tell which roof-plane measurement belongs to which image. The equipment label is blurred. A note says “tree on west side,” but no photo shows its position. The assessor may have seen everything needed on site. The handoff did not preserve what they knew.
The resulting messages are often treated as normal collaboration. “Which dimension is this?” “Can you get a closer panel photo?” “Is the vent staying?” Each question is reasonable, but the designer has quietly become the owner of survey completion. Work is interrupted, answers arrive outside the project record, and the eventual design may depend on a detail nobody marked as provisional.
Preventing that chase does not require collecting every conceivable fact at every visit. It requires a sender-owned evidence package, a receiver-owned acceptance decision, and a defined path for gaps. The requested output determines what “enough” means. An early internal concept, a survey-driven customer layout, and a later technical handoff should not use the same evidence threshold.
This guide covers the field-to-design control. For the inspection subjects themselves, use the solar site survey checklist. For the wider qualification path before a request reaches design, use the solar project intake process. This article does not provide site, safety, structural, electrical, engineering, permitting, utility, legal, financial, or code advice. Qualified reviewers and the applicable authorities still govern each live project.
Why do designers end up chasing missing site information?
Designers chase site information when a team equates “uploaded” with “usable.” The survey package may contain many files but lack identity, context, evidence quality, or an owner for gaps. Without a receiving rule, design must reconstruct the visit, ask private questions, or proceed with assumptions that later readers cannot see.
The operational failure occurs between capture and use. A field assessor experiences the property as one continuous scene. They can look from a roof plane to a nearby obstruction, point at equipment, and remember why a measurement was taken. The designer receives fragments. Photos, notes, sketches, customer statements, drawings, and dimensions only become design evidence when their relationships survive the handoff.
The U.S. Department of Energy’s photovoltaic system design overview explains that a PV system may include modules, mounting structures, power electronics, and storage. That system context helps explain why a generic “site surveyed” status is weak. Site, structural, electrical, equipment, energy, and documentation inputs can affect different parts of the work. The page is general technical information, not a project acceptance rule.
Separate six conditions that can all look like “missing information” in a message thread:
| Condition | What the designer actually has | Correct operational response | What not to do |
|---|---|---|---|
| Absent | No record addresses the required subject | Return it, commission focused capture, or limit the output under policy | Ask the designer to hunt through unrelated files |
| Unidentifiable | A photo, measurement, or note exists but cannot be tied to a site feature | Send it back for indexing or confirmation by the person who captured it | Guess from upload order or filename |
| Unreadable | The intended record exists but the needed detail cannot be inspected | Re-capture it or obtain an alternate source | Treat the checkbox as proof that the detail was readable |
| Conflicting | Two plausible records disagree | Freeze the affected field and route a documented resolution | Choose whichever value makes the design easier |
| Provisional | The team knows the value or condition is not yet verified | Permit only the work defined for that evidence state | Let the provisional input flow into an unqualified customer-facing output |
| Changed | The package was accepted, then the site, scope, or source record changed | Reopen affected work and identify downstream artifacts | Replace a file silently and assume prior work remains valid |
More mandatory fields will not solve these conditions by themselves. A person can enter “see photos” in a required note, upload an unreadable label, or check “measured” without showing how the dimension maps to the roof. The control must describe what acceptable evidence looks like and what happens when it is not present.
This is also why the closest inventory page, why two designers can produce different yield forecasts, solves a later problem. That guide reconciles model inputs and reporting boundaries after designers have a basis to work from. Here, the question is whether the captured site package can enter design at all.
Which seven controls stop the missing-information chase?
Use seven connected controls: sender attestation, stage-specific evidence rules, indexed photos and measurements, field-level acceptance tests, a receiver decision, owned exception tickets, and change-triggered reopening. Together they move clarification from private designer follow-up into a visible workflow where the right role captures, accepts, limits, resolves, and updates the site evidence.
1. Make the sender attest to the package
The person submitting the site package should confirm more than “survey complete.” Ask that sender to attest that the records belong to the named project, required subjects were addressed, files are readable, measurements and photos are indexed, known conflicts are disclosed, and remaining gaps are listed. This makes the handoff an accountable action rather than the end of an upload.
Sender ownership does not mean the assessor must decide whether the evidence supports every design task. The assessor owns the integrity of the capture record. The receiving designer owns whether that record supports the requested work. Keeping those authorities separate avoids two bad outcomes: design becoming the default data collector, or field staff declaring technical sufficiency for a downstream deliverable they do not control.
Use a named person or controlled role, not “field team.” If several people contributed, one submitter still needs to review the assembled package. That person can route questions back to contributors before the package reaches design.
A simple attestation can read: “I confirm that these records belong to project SP-1047; required capture items for the survey-driven layout stage are addressed or listed as exceptions; photos and measurements use the attached index; and no known site change after capture is omitted.” Adapt the wording to your policy, permissions, and project types.
2. Set evidence rules for the requested design stage
Do not maintain one universal definition of complete. Write the output first, then decide what evidence it needs. A preliminary internal roof-fit discussion may permit current imagery with geometry labelled provisional. A survey-driven layout may require mapped roof dimensions and readable obstruction records. A later electrical or engineering workflow needs its own qualified review and jurisdiction-specific inputs.
The shared definition of a complete solar design request covers that receiving agreement in depth. At the field handoff, turn the agreement into a matrix the assessor can use before submitting.
| Evidence group | Early internal concept | Survey-driven layout | Later technical handoff |
|---|---|---|---|
| Site identity | Confirmed address and project record | Confirmed identity on every indexed capture set | Reconciled identity across current controlled records |
| Roof geometry | Documented imagery or labelled planning geometry if policy permits | Indexed dimensions, plane identities, and visible measurement basis | Current records reviewed for the specific technical release |
| Obstructions and shade context | Known major conditions labelled | Mapped subjects with photos, position, and relevant dimensions | Review appropriate to the model, design, and jurisdiction |
| Electrical evidence | Excluded or clearly marked pending when irrelevant to the concept | Captured to the team’s defined survey scope, with readable identities where required | Reviewed by the responsible qualified role for the intended output |
| Customer or site statements | Named speaker, date, and exact statement | Same, plus affected decision and verification trigger | Replaced or supplemented by the source required for technical reliance |
| Open gaps | Visible and excluded from use | Must follow a return or limited-work rule | Must follow the release authority’s rule |
These are operating examples, not universal solar requirements. A team should define its stages around the work it actually performs and the authorities it must satisfy. The useful test is behavioral: can the assessor predict whether a gap will cause return, limited work, alternate work, or escalation?
NASA’s requirements-management guidance applies to NASA programs rather than solar projects. It discusses identifying, tracing, controlling, and managing changes to requirements. The transferable lesson is narrow: a downstream task works better when the expected output and its governing inputs can be traced and changed deliberately.
3. Index every photo and measurement by subject
An image folder is not a photo index. The designer should be able to identify what each file shows, where the subject is, which direction or viewpoint applies, and why the file matters. Use a consistent project identifier, subject, orientation or location, sequence, and capture date in the filename or managed metadata.
For example, SP1047_main-service_exterior_01_2026-08-30.jpg is more useful than IMG_8841.jpg. SP1047_roof-plane-N2_eave-span_2026-08-30.jpg can point to a measurement record with the same plane ID. The exact syntax matters less than consistency and uniqueness.
The index should answer questions the filename cannot carry cleanly:
- What site feature is visible?
- Where is it relative to the site or roof-plane map?
- What orientation or viewpoint was used?
- Which measurement, note, or checklist item does it support?
- Is the subject existing, proposed, stated, assumed, or unresolved?
- Is any detail intentionally redacted or outside the capture scope?
Apply the same discipline to dimensions. “22 ft” is not a measurement record. Name the endpoints, plane or surface, method, unit, and any limitation that matters to later interpretation. A dimension sketched on a roof map should use the same plane IDs as the photo set. If the assessor could not safely or reliably capture an item, record that condition as an exception rather than inventing precision.
A prettier folder is beside the point. The index must make the assessor’s field context inspectable without requiring them to narrate the entire visit again.
4. Give each required field an acceptance test
Required fields should specify the evidence condition that counts as usable. “Panel photo required” is incomplete. Does design need the equipment context, a readable label, conductor visibility, clearances, or merely confirmation of location for the requested stage? Those may require different images and different reviewers. The field rule should state the subject, evidence form, readability test, stage, and failure path.
Write tests in observable language. “Good photo” invites opinion. “The full equipment label is in frame and the characters needed by the receiving role are legible at normal review zoom” can be checked. “All roof measurements complete” is vague. “Every plane in the roof map has the dimensions required by the stage matrix or a linked exception” connects completeness to a defined set.
Use five evidence states consistently:
- Documented: an identifiable source record is attached and usable within its stated limits.
- Stated: a named person provided the information, but it has not been independently verified.
- Assumed: the team introduced a planning condition for defined limited work.
- Missing: the required evidence is absent, unidentifiable, or unusable.
- Conflicting: current-looking records disagree and need resolution.
These labels are a proposed operating vocabulary, not an external standard. They matter only if they change behavior. If “missing” and “documented” both allow the same output without a visible qualification, the workflow is decorating the record rather than controlling it.
NASA’s technical-data management guidance discusses planning how data are acquired, identified, accessed, managed, protected, and used in NASA work. Solar companies need their own security, privacy, retention, and technical policies, but the cross-domain point is useful: possession of a file does not establish its identity, authority, or permitted use.
5. Let design accept, limit, or return the package
The receiving designer needs more than a “complete/incomplete” button. Give the receiver three outcomes tied to the requested output:
- Accept: The package meets the evidence rules for the named stage, subject to normal design review.
- Accept for limited work: Specific tasks may begin while exceptions remain visible, and prohibited downstream uses are recorded.
- Return or redirect: The package cannot support the requested task; the sender must correct it, collect focused evidence, change the deliverable, or route it to another qualified role.
Record the receiver, timestamp, package version, stage, decision, and exceptions. Acceptance is not approval of the future design. It only says the input basis supports beginning the defined work. The solar design review checklist governs a later release decision after an output exists.
This receiving authority is essential. If a manager tells designers to “just start” regardless of evidence, every mandatory field becomes advisory. Designers will either guess or preserve their own private return rules. Different people then accept different risk under the same status label.
Conditional acceptance should be narrow. It can be useful when unaffected work is genuinely separable, such as organizing the project record while one roof-plane dimension is being recaptured. It is not permission to hide uncertainty. The resulting task and any output must retain the exception, affected scope, and prohibition on unsupported downstream use.
6. Convert every gap into an owned exception ticket
A missing-data message asks a question. An exception ticket controls work. It should identify the missing or conflicting item, why it matters, the requested evidence, owner, next action, due event, permitted work, prohibited use, and reopening trigger. Link the ticket to the exact package version and affected design task.
Assign ownership to the role that can resolve the gap. That may be the site assessor, scheduling coordinator, customer contact, sales representative, survey vendor, designer, project manager, or qualified technical reviewer. The designer can describe the dependency without owning the field trip, customer follow-up, or policy decision.
A focused evidence request does not authorize hazardous access. State the capture role, permitted access and current safety controls; use an alternate record or qualified route when direct capture is unsafe or outside the assigned task. Never ask the homeowner to remove covers or manipulate electrical equipment to satisfy a missing field.
Avoid a generic “need more photos” ticket. State the subject and acceptance test: “Capture one full-frame image of the main service equipment label; characters required by the electrical reviewer must be readable; link it to equipment ID E-01.” A focused request is easier to perform and easier to accept.
NASA’s interface-management guidance concerns NASA system interfaces, responsibilities, and interactions, not solar survey practice. As a bounded analogy, it reinforces the value of naming what crosses a role boundary, who controls it, and how a discrepancy is resolved.
7. Reopen affected work when site information changes
The chase can return after acceptance if new files silently replace old ones. A newer utility record, changed equipment preference, reroof decision, removed tree, service upgrade, revised building plan, or corrected dimension may affect work already underway. Define which changes reopen which tasks and who decides the impact.
Do not overwrite the accepted package without history. Retain the prior version, identify the changed field and source, record who submitted it, and ask the responsible receiver to assess affected outputs. Layout, shading, energy modeling, electrical workflow, bill of materials, financial scenarios, and proposal content may not all need the same response.
NASA’s configuration-management guidance describes identifying configurations, controlling changes, tracking status, and auditing configuration information for NASA programs. It does not impose a solar workflow. The transferable principle is that a current state and its changes should remain identifiable when multiple work products depend on them.
The solar design source-of-truth guide covers longer-lived field and artifact authority across the project. The site package remains one controlled input to that wider system. Its accepted version should not compete with a newer unreviewed file in a chat thread.
Map the Accepted Site Package Into Design
Explore how a controlled project basis can connect to solar modeling, layout, equipment outputs, and proposal work while your team retains evidence and review authority.
Explore Commercial Solar DesignHow should a solar team implement these controls?
Start with one recent clarification chain, identify the missing control behind each question, and define a stage-specific pilot. Build the smallest evidence matrix that changes a receiving decision, then test it on live packages. Review return reasons, conditional work, reopened tasks, and unused fields before expanding the workflow across project types.
Use this numbered implementation process:
- Reconstruct one chase. Place the original package, designer questions, replies, revisions, and final output in time order. Mark where knowledge existed but failed to cross the handoff.
- Name the requested stage. Choose one repeatable deliverable for the pilot. Do not try to standardize every survey, customer type, and technical release at once.
- List decision-changing subjects. Include only site information whose absence, ambiguity, or conflict changes acceptance, permitted work, review ownership, or output use.
- Define an evidence test for each subject. State what must be visible, identifiable, mapped, dated, or attributed, plus which evidence states are permitted.
- Build the photo and measurement index. Reuse subject IDs across maps, files, notes, and exception tickets. Test whether a designer who missed the visit can navigate it.
- Assign sender, receiver, and exception owners. One person attests to the package, one role makes the intake decision, and each gap routes to someone able to resolve it.
- Pre-write the three receiving outcomes. Define accept, limited work, and return or redirect for the pilot stage. Include prohibited uses and reopening triggers.
- Run a bounded live pilot. Keep the old workflow available if required by risk policy, but record which route each package used and why.
- Review behavior, not form completion. Look at clarification topics, return reasons, ageing exceptions, reopened work, and fields designers never consulted. Do not treat a lower return count as proof of improvement without checking enforcement.
- Revise and extend deliberately. Remove low-value fields, tighten ambiguous tests, preserve rule versions, and add another design stage only after the pilot is understood.
Do not set an external target for acceptable return rates or clarification volume unless you have evidence applicable to your workflow. An internal baseline can still help. Define the unit first. One “clarification” might mean a message, a topic, an exception ticket, or a returned package. Pick one definition and keep it stable before comparing periods.
The implementation meeting should include field and design roles. Managers can approve the operating policy, but the people who capture and inspect the records will expose ambiguous labels quickly. Ask an assessor to demonstrate how they would submit a complex site. Ask a designer to find one roof measurement, one obstruction, and one equipment record without coaching.
Copy-ready site-package acceptance record
Copy this asset into a form, ticket template, or controlled project record. Adapt it to your project stages, technical responsibilities, permissions, and jurisdictions.
SITE-PACKAGE ACCEPTANCE RECORD
Project ID:
Site address or controlled site ID:
Requested design stage:
Requested output:
Intended audience and decision:
Package version:
Capture date(s):
SENDER ATTESTATION
Submitting person:
Capture contributors:
Required evidence matrix version:
Project identity checked: yes / no
Files readable at normal review zoom: yes / no / exception linked
Photos and measurements indexed: yes / no / exception linked
Known conflicts disclosed: yes / no
Changes after capture disclosed: yes / no / unknown
EVIDENCE REGISTER
Subject ID:
Subject or field:
Evidence state: documented / stated / assumed / missing / conflicting
Source file or record:
Capture source and date:
Location, orientation, or endpoints:
Supported design task:
Known limitation:
OPEN EXCEPTIONS
Exception ID:
Missing or conflicting item:
Why it affects the requested work:
Resolution owner:
Requested evidence or decision:
Next action and due event:
Work permitted while open:
Use prohibited while open:
Escalation trigger:
RECEIVING DECISION
Decision: accept / accept for limited work / return or redirect
Receiving person:
Decision timestamp:
Accepted package version:
Permitted work:
Prohibited use:
Linked exceptions:
Required reviewer or approval path:
CHANGE CONTROL
Changed field or record:
Old and new package versions:
Change source and date:
Affected work products:
Impact reviewer:
Reopen / no impact / replace decision:
Decision record and timestamp:
The asset intentionally separates evidence capture from receiving judgment. Do not collapse the sender attestation and receiver decision into one signature unless the same authorized person legitimately performs both roles under your policy.
What should happen when site evidence is incomplete?
Incomplete evidence should trigger a preselected route, not an improvised message chain. Return the package when the requested work cannot proceed, authorize limited work when unaffected tasks are separable, redirect to focused capture or qualified review, and escalate policy conflicts. Keep the gap, owner, permitted work, prohibited use, and resolution visible.
Use the impact on the requested output to choose a route:
| Decision route | Use when | Required record | Closure condition |
|---|---|---|---|
| Return for correction | The sender can repair indexing, identity, readability, or a required field before design begins | Failed acceptance test, sender, exact correction, package version | Corrected package is resubmitted and accepted |
| Focused recapture | The site must be revisited or another source obtained for a specific subject | Capture request, responsible role, access or safety boundary, affected task | New evidence passes the field test |
| Limited work | The missing item does not affect a separable task and uncertainty can remain visible | Permitted task, prohibited use, assumption if any, expiry and reviewer | Gap closes or limited output is retired or superseded |
| Alternate deliverable | Available evidence supports a different decision than originally requested | New output, audience, limits, and approval | Receiver accepts the revised request |
| Qualified escalation | The gap involves safety, engineering, code, authority, utility, contractual, or other controlled judgment | Question, sources, jurisdiction, responsible reviewer | Authorized decision is recorded |
| Policy escalation | Sender and receiver disagree about the rule itself | Competing interpretations, affected work, policy owner | Rule is clarified without forcing a one-off hidden exception |
Illustrative example, not a customer case
A residential survey package is submitted for a survey-driven customer layout. Roof-plane photos and mapped dimensions are identifiable. The photo index lists the main service equipment, but the label image is blurred. A note also says a mature tree west of the home may be removed, with no confirmation or schedule.
The designer does not send two unrelated messages and continue silently. The receiver records two exceptions. The unreadable equipment label routes to focused recapture under the team’s electrical-evidence rule. The tree statement is labelled “stated,” attributed to the homeowner and capture date, and linked to a verification trigger.
The stage matrix permits roof-model and array-layout exploration using the mapped geometry. It prohibits electrical release and prohibits presenting a tree-removal scenario as the verified site condition. The receiver conditionally accepts only the separable work. The output keeps existing-tree and possible-removal scenarios distinct if the team’s policy allows that comparison.
When a readable label arrives, the electrical exception routes to the responsible reviewer. If written confirmation about tree removal later changes the accepted site basis, the workflow opens change control and asks which design and proposal artifacts require review. No one needs to remember which chat contained the update.
This example does not establish that the site is suitable, the tree can or should be removed, the electrical condition is acceptable, or any design will receive approval. Those conclusions belong to the responsible project roles and external authorities.
Audit the chase without blaming the chaser
Designer questions can reveal a weak control, but do not treat every question as waste. A skilled designer should challenge ambiguous or conflicting evidence. The target is avoidable reconstruction, not responsible review.
Classify clarification events by mechanism:
- required subject was absent;
- evidence existed but lacked identity;
- file was unreadable;
- field rule was ambiguous;
- evidence conflicted;
- requested output changed;
- designer asked for information outside the agreed stage;
- exception owner did not act;
- downstream work used a package that had not been accepted.
Then improve the right control. Repeated unidentifiable photos call for a better index, not necessarily another mandatory photo. Frequent requests outside the stage may require designer training or a revised service definition. Long-open exceptions may signal ownership or scheduling problems. A package returned for a genuine changed condition is not the same failure as a form that never captured the subject.
Where should software support the site-to-design handoff?
Software should preserve project identity, indexed evidence, package version, stage, sender attestation, receiver decision, exceptions, assumptions, and affected outputs in one reviewable path. It can enforce fields and link work products, but it cannot verify every site fact, make engineering judgments, interpret every jurisdiction, or replace responsible external approvals.
Configure the operating rule before automating it. A mobile form can prompt an assessor to capture a subject. File naming can preserve identity. Status rules can prevent unaccepted packages from entering designated work. Notifications can route exceptions. Version records can show which evidence supported a design. None of those features decides what your project stage requires or whether a source is truthful.
SurgePV supports 3D roof modeling, solar array layout, shading analysis, energy-yield and financial modeling, electrical workflow, bill-of-materials output, and proposal generation. Results depend on source data, assumptions, equipment models, configuration, and review. SurgePV does not replace site verification or approval by the responsible engineer, authority, lender, insurer, or utility.
The solar designing workspace belongs after the team has accepted a controlled project basis. Opening a model does not cure an unidentifiable photo, resolve a conflicting measurement, or transfer the assessor’s capture responsibility to the designer.
The practical software boundary is clear. A system can keep roof-plane IDs beside the model, but a person must decide whether the mapped dimensions are usable. It can connect a layout to an energy scenario and proposal, but the team must preserve provisional inputs and review changes. It can support an electrical workflow, but qualified roles and applicable authorities retain their responsibilities.
Build the handoff so the designer can answer five questions without messaging the assessor: Which site and package version am I using? What output was requested? Which evidence supports each relevant subject? What remains uncertain? Who owns each open exception? When those answers are visible, designer follow-up becomes targeted review rather than routine archaeology.
Frequently Asked Questions
Who owns missing information after a solar site survey?
The role assigned to capture or obtain the information should own the open item until it is resolved or formally waived by an authorized reviewer. The designer should identify why the gap blocks or limits work, but should not become the default collector. A named exception owner, next action, due event, and affected output keep ownership visible.
Should a designer start when some site information is missing?
Only when a pre-agreed rule allows limited work and the output can preserve the uncertainty. The acceptance record should name the missing item, permitted task, prohibited use, assumption if any, and verification trigger. If the gap could change the requested result or create an unsafe or unsupported conclusion, route it for qualified review instead of guessing.
How should solar site-survey photos be named?
Use a project identifier, location or subject, orientation or viewpoint, sequence, and capture date, then connect each file to a photo index. A good name helps a designer distinguish items such as the main service panel exterior, equipment label, north roof plane, and an obstruction. Never rely on upload order as the only explanation.
Is a complete site-survey checklist enough for design?
No. A checked box confirms that someone interacted with a field, not that the evidence is identifiable, readable, current, or adequate for the requested design stage. Pair the checklist with acceptance criteria, source and date labels, a photo and measurement index, exception behavior, receiver sign-off, and rules for reopening work when site information changes.
Where can SurgePV support the survey-to-design handoff?
SurgePV can support 3D roof modeling, array layout, shading analysis, energy-yield and financial modeling, electrical workflow, bill-of-materials output, and proposal generation. Results depend on source data, assumptions, equipment models, configuration, and review. The software does not replace site verification or approval by responsible engineers, authorities, utilities, lenders, or insurers.
Review Your Site-to-Design Workflow in a Guided Demo
Bring one project stage and its evidence rules to see how controlled inputs can connect to modeling, layout, equipment outputs, and proposal work.
Book a Guided DemoSources
Primary research and reference material used for this desk-research article.
Where this fits
This article is part of SurgePV's Solar Business & Operations hub, which works through the topic from first principles to the decisions a project team actually has to make.


