The Photo Documentation Order Adjusters Actually Want on a Roof Claim

An insurance adjuster does not look at your roof photos the way you do. You climbed the roof, you found the damage, you know it's there. The adjuster opens a folder of forty-seven JPEGs on a laptop, often without ever standing on the roof, and has to reconstruct the whole inspection from those images alone. If your photos don't tell the story in the order their desk software and their training expect, the claim slows down, gets partially paid, or comes back as a reinspection request three weeks later.

The photos you take are not the documentation. The order you take them in is the documentation. A pile of close-ups of damaged shingles proves nothing on its own — an adjuster can't tell which slope they came from, whether that's hail or blistering, or whether the rest of the roof matches. A sequence that walks from the street to the address to the slope to the test square to the single mark, with each step locating the next, turns the same photos into something a desk reviewer can approve without a second site visit.

What follows is the photo order that works, slope by slope, shot by shot, with the naming and the edge cases that keep claims from bouncing. It's built from how adjusters actually review files, not from how it feels to walk a roof.

Why the order matters more than the photos

Most roofing companies lose money on claims documentation in the same two places: reinspections and supplements that never get approved. Both usually trace back to a photo set that's complete but unreadable. The damage is in there somewhere. The adjuster just can't follow it.

Think about what the desk reviewer is doing. They have a scope of loss — a line-item estimate — and a folder of photos. Their job is to match every line they're being asked to pay for to a photo that justifies it. Tear-off and re-felt on the north slope? They need a photo that clearly shows the north slope and clearly shows storm damage on it. Drip edge? They need a photo of the existing drip edge. Ridge cap? Same. If they can't find the photo, they don't pay the line. They don't call you to ask. They just deny or defer it, and the burden bounces back to you to supplement.

When your photos run in a logical order — wide to tight, ground to roof, address to slope to mark — the reviewer can scroll once and check every box. When they run in the order you happened to shoot them (a blistered vent, then the front elevation, then a hail strike, then your truck), the reviewer has to hunt, and hunting is where claims die.

There's a second reason order matters: chain of evidence. An adjuster, and behind them a claims manager and sometimes an attorney, wants to believe these photos came from this roof on this day and that the damage is real and recent. A sequence that starts at the street with the house number, moves to the elevations, then onto the roof, then down to individual marks, builds that chain visually. A folder of disembodied close-ups builds nothing. It could be any roof, any storm, any year.

Get the order right and two things happen. Reinspection rates drop because the file is self-explaining. Supplement approval improves because every line you write already has a photo sitting next to it.

The mental model: five zoom levels

Every good roof claim photo set moves through five levels of zoom, in order, and then repeats the tightest levels for each slope and each item. Hold this model in your head and the sequence becomes automatic.

1. Locate — Where is this house? Street, house number, full address context.
2. Orient — What does the building look like? Four elevations, roof geometry, story height.
3. Establish the slope — Which roof plane am I on, and how does it sit relative to the rest of the roof?
4. Establish damage on the slope — A test square or marked area showing the density of damage on that plane.
5. Prove the single mark — One hailstrike or wind tear, in focus, with scale and circle.

Levels 1 and 2 you do once per property. Levels 3, 4, and 5 you repeat for every slope. The discipline is to never jump from a wide shot straight to a single mark without the steps in between — because the steps in between are what tell the adjuster the mark belongs to a slope, the slope belongs to the roof, and the roof belongs to the address.

Most contractors shoot level 5 obsessively and skip levels 3 and 4. That's exactly backward. The adjuster can usually see the single mark fine. What they can't see is the density and the location, and density and location are what justify a full replacement instead of a repair.

The exact photo order, shot by shot

Here is the full sequence for a standard hail and wind inspection on a single-family home. Adapt the slope count to the roof, but keep the order.

Phase 1 — Locate the property (3 to 5 photos)

• House number / address. A clear shot of the numbers on the house or mailbox. This is the first photo in the folder and it anchors everything. If the number is faint, shoot the mailbox and the door number both.
• Street context. One photo from the street showing the front of the house in its setting. This proves a real address, not a staged shot.
• Date reference (optional but strong). A newspaper, a phone showing the date, or simply rely on EXIF — but if your carrier or program strips metadata, a visible date reference photo at the front is cheap insurance.

The purpose here is purely chain of evidence. It costs you ninety seconds and it preempts the single most corrosive adjuster suspicion: that the photos aren't from this property or this date.

Phase 2 — Orient the building (4 to 8 photos)

• Four elevations. Front, right, left, rear — one clean photo of each side of the house. These do double duty: they show siding, gutters, window screens, and soft metals that corroborate the hail story, and they give the adjuster the roof geometry from the ground.
• Collateral damage on elevations. If you see dented gutters, torn screens, damaged window wraps, spatter on the deck or AC fins, dings on the garage door — shoot them now, framed so the adjuster can see they're on the house you just located. Collateral damage on soft metals is often the strongest corroboration of a hail event because it can't be faked by foot traffic or manufacturing defect.
• Roof overview from the ground or a drone. A wide shot showing the overall roof shape and the number of slopes. This is the map the adjuster will use to place every later photo.

The elevations matter more than contractors think. When a desk reviewer is on the fence about whether an event was severe enough to total a roof, dented gutters and bruised AC fins on the elevation photos are what push them over. They also date the storm: fresh, shiny dents in aluminum read as recent; old oxidized dents read as prior.

Phase 3 — Establish each slope (1 to 2 photos per slope)

Now you're on the roof. Before any close-up, shoot the slope as a whole.

• Slope overview. Stand at the eave or ridge and shoot the entire plane so the adjuster sees which slope this is and its general condition. Frame a reference — a chimney, a vent, a valley, the ridge — so this photo can be tied back to the roof overview from Phase 2.
• Slope direction marker. Many pros lay a directional chalk note or a small dry-erase board reading "NORTH SLOPE" in the frame, or use an app that overlays slope and compass. This single habit eliminates the most common reinspection trigger: the adjuster can't tell which slope a close-up came from.

Do this for every slope before you start hunting marks. The sequence in the folder should read: north slope overview, then north slope test square, then north slope marks — and only then move to the next slope. Never interleave slopes. A folder that jumps north, east, north, west, east is unreadable.

Phase 4 — The test square (3 to 4 photos per slope)

The test square is the single most important documentation move on a hail claim, and it's the one most often done sloppily. A test square is a marked 10-foot by 10-foot area (100 square feet) in which you count and circle every storm strike. It converts "there's damage" into "there are X hits per square," which is the language carriers use to decide repair versus replace.

• Test square — marked and bounded. Chalk or tape out the 10x10 area. Shoot it wide so the boundary is visible and the adjuster can see the size of the area you're counting in. Put a slope marker or chimney in the frame so it ties to the slope overview.
• Test square — hits circled. Walk the square, circle each strike with chalk or a lumber crayon. Shoot the whole square again with all hits circled. The adjuster now sees density at a glance.
• Test square — hit count callout. Many programs want a number written in the frame: "NORTH — 9 HITS / 100 SF." Write it on your board or chalk it on the roof and shoot it.

A few rules that separate clean test squares from ones that get challenged:

• Make the square representative, not cherry-picked. If you mark your square in the one cluster of damage on an otherwise clean slope, an adjuster who reinspects will catch it and trust nothing else in your file. Put the square where the damage density is typical for that slope.
• Count consistently. Decide what counts as a hit (a bruise that fractures the mat, granule loss exposing asphalt, a soft spot you can feel) and apply it the same way every square. Don't count blisters, don't count manufacturing flaws, don't count foot traffic. Adjusters know the difference and inconsistency reads as inflation.
• One square per slope, minimum. Different slopes face different directions and take hail differently. A square on each slope shows the directional pattern that's the fingerprint of a real storm.

Phase 5 — Prove the single mark (3 to 6 photos per slope)

Now, and only now, the close-ups. These prove the nature of the damage — that it's hail or wind, not age, not blistering, not a defect.

• Circled hit with scale. Circle one clear strike with chalk and place a coin, a chalk-line gauge, or a tape measure next to it for scale. Shoot it in focus. A quarter is the field standard for scale because everyone recognizes its size.
• The same hit, tighter. Move in for a macro of the same circled strike showing the fractured mat, the bruise, the granule displacement. This is where you prove it's impact, not weathering.
• Hail bruise on the underside / soft spot. If your program supports it, a photo showing the spongy give of a fresh bruise (often done as a video clip or a thumb-press shot) reinforces freshness.
• Directional / wind damage. For wind: creased shingles, lifted tabs, torn shingles, missing shingles with the felt or deck exposed. Shoot the crease line clearly — a wind crease is a stress fracture across the shingle and is unmistakable when lit and framed right.
• Repeat for 3 to 4 representative marks per slope. Not fifty. Three or four clean, circled, scaled, in-focus marks per slope beat fifty blurry ones. Volume of bad photos reads as noise; a few sharp ones read as evidence.

Phase 6 — Components and penetrations (as present)

After the field of the slope, document every component you intend to claim, because each is a separate line the adjuster has to justify.

• Ridge and hip caps — overview and a damaged close-up.
• Valleys — metal or woven, with any damage.
• Drip edge / rake edge — existing condition; you can't get it paid if there's no photo of it.
• Pipe jacks / boots — cracked or damaged rubber, sun rot.
• Vents, turbines, ridge vent — dents, dislodgement.
• Flashing — step, counter, chimney, sidewall.
• Skylights — cracked glass, damaged frames.
• Gutters and downspouts — dents, separation (ties back to elevation shots).
• Detached structures — sheds, detached garages, patio covers, often forgotten and often covered.

Each of these is a photo the adjuster needs to pay the corresponding line. The pattern is the same as the slopes: a locating shot, then the damage.

Two components get underdocumented constantly and cost real money. The first is drip edge and rake edge existing condition. Even when there's no storm damage to the drip edge itself, code in most jurisdictions requires drip edge on a re-roof, so the line gets written — but an adjuster won't pay to replace something they can't see exists. A plain photo of the existing drip edge on each eave and rake is the entire justification. Skip it and you're supplementing for a sixty-dollar line three weeks later. The second is starter course and ridge ventilation. If the roof has a ridge vent, photograph it; the replacement line depends on the adjuster seeing the existing system. If you intend to claim starter strip, show the existing starter at a lifted first course. These are small lines individually, but on a full re-roof they add up, and they're the ones that quietly fall off scopes for lack of a photo.

A useful rule: for every line you plan to write on the estimate, ask "which photo proves this?" before you come off the roof. If you can't name the photo, go take it. The estimate and the photo set should be built as a matched pair — every line has its picture, every picture earns its line.

Phase 7 — Measurements, layers, and the deck

• Shingle layers at a cut. A photo of a tab lifted or cut showing the number of layers and the deck type. Layers affect tear-off cost; deck type (plank versus OSB versus plywood) affects re-deck lines.
• Underlayment and deck condition where visible.
• Roof complexity markers — multiple stories, steep pitch (a pitch gauge in frame), cut-up roofs. These justify steep and difficult charges.
• Overview measurements / diagram if you're using an aerial measurement report, reference it; if you hand-measured, photograph the roof with dimensions noted.

A field workflow you can actually run

Knowing the order isn't enough if the crew can't execute it consistently on a hot roof with a homeowner watching. Here's the workflow that makes the order stick.

Before you climb (5 minutes on the ground). Shoot Phase 1 and Phase 2 from the ground first — address, street, four elevations, collateral damage, roof overview. Doing these first means that even if you have to come off the roof early (weather, a hostile homeowner, a safety call), you already have the locating and orienting half of the file.

On the roof, one slope at a time. Pick a starting slope and complete Phases 3, 4, and 5 for it — overview, test square, marks — before moving to the next slope. Resist the urge to wander to the "good damage" on another slope. Finish the slope you're on. This is the single discipline that keeps the folder readable.

Narrate as you go. If your documentation app supports voice notes or captions, say which slope and what you're shooting as you shoot. "North slope, test square, nine hits." It speeds your later sorting and it gives the desk reviewer context.

Components last, then layers and deck. After all slopes, sweep the components (Phase 6), then do your cut for layers and deck (Phase 7). Doing the destructive cut last means you've documented everything intact first.

Sort and name before you leave the driveway. The five minutes you spend in your truck renaming and ordering photos saves an hour of confusion at the desk and prevents the reinspection that costs you a day. More on naming below.

A realistic shot count for a clean single-family hail claim runs 35 to 60 photos: 4–5 locating, 6–8 orienting, then per slope roughly 2 overview + 3 test square + 4 marks = 9, times four slopes = 36, plus 10–15 components and cuts. If you're under 25 photos, you're almost certainly thin on test squares or slope overviews. If you're over 100, you're probably drowning the adjuster in redundant close-ups and burying the signal.

Naming and ordering so the folder reads itself

The best photo set in the world fails if it lands on the desk as IMG_3041 through IMG_3097 in random capture order. Two adjusters out of three will not reorder your folder for you. Make the filenames do the work.

A naming scheme that sorts correctly and reads clearly:

01_address_house-number
02_street_context
03_elevation_front
04_elevation_right
05_elevation_rear
06_elevation_left
07_collateral_gutter-dents
08_roof_overview
10_north_slope-overview
11_north_test-square-bounded
12_north_test-square-circled
13_north_hit-scale-quarter
14_north_hit-macro-mat-fracture
20_east_slope-overview
21_east_test-square-bounded
...
50_ridge-cap-damage
51_drip-edge-existing
52_pipe-boot-cracked
60_shingle-layers-cut
61_deck-osb

Number the prefixes with gaps (10, 20, 30 per slope) so you can insert a photo later without renumbering the whole set. Lead with a zero-padded number so the folder sorts in capture order, not alphabetical chaos. Put the slope direction in the name as well as the description, so a reviewer scanning filenames can see the slope coverage without opening a single image.

If your CRM or claims software assigns photos to scope lines, name to their convention, not yours. The goal is for the adjuster to map filename to line item without thinking. Every second you save the reviewer is a second they're not spending looking for a reason to defer.

Keep the originals. Whatever you submit, archive the untouched files with intact EXIF metadata. If a claim goes to appraisal or litigation, the original timestamped, geotagged files are your strongest evidence, and a copy that's been through three apps and stripped of metadata is far weaker.

Where roof selection and storm modeling fit before you ever climb

Everything above assumes you're already on the right roof. A surprising amount of wasted documentation comes from climbing roofs that were never going to produce a payable claim — roofs the storm missed, or roofs so young the damage won't total — while missing roofs in the same neighborhood that took the worst of it. The cleanest photo workflow can't save a claim on a roof that didn't get hit hard enough to replace.

This is where pre-climb intelligence changes the economics. RoofPredict ranks addresses by two things that decide whether a documentation visit is worth a crew's time: a per-roof age range estimated from aerial imagery, and storm physics modeled per individual roof rather than per zip code or per radar cell. Instead of canvassing a whole subdivision and climbing every roof, you point crews at the doors where an aging roof and a modeled storm hit overlap — the roofs most likely to both have damage and be near the end of their service life, where a full inspection and a clean photo set are most likely to result in an approved replacement.

Be clear about what this is and isn't. The age output is a range, not a manufacture date or an install certificate — it tells you a roof reads as roughly 14 to 18 years old, not that it was installed in a specific month. The storm model is odds, not proof: it tells you which roofs most likely took the brunt of a given event so you can prioritize who to inspect first. It does not document damage, it does not establish causation, and it is never a substitute for the test square and the circled hit. The adjuster still needs your photos. What the modeling does is make sure the photos you go take are on roofs worth documenting — it sharpens the route, not the claim. The evidence that gets a claim paid is still made on the roof, in the order described here.

Used honestly, that's a real edge: fewer wasted climbs on roofs that won't total, more crew hours spent documenting the roofs where a complete, well-ordered photo set actually converts. The contractor still documents conditions and writes the estimate; the carrier still decides coverage; the homeowner still owns the claim. The intelligence just tells you which doors to knock first.

What pros get wrong (and how the desk reads it)

The difference between a file that pays and a file that bounces usually comes down to a handful of recurring mistakes. Here's what they are and how an adjuster interprets each.

Mistake 1 — All marks, no map

Fifty close-ups of hail hits, zero slope overviews, zero test squares. How the desk reads it: "I can see damage but I can't tell where it is or how dense it is. I'll approve a repair, not a replacement, and let them supplement if they disagree." The fix is Phases 3 and 4 — overviews and test squares — which you probably skipped because the marks felt more important.

Mistake 2 — The cherry-picked test square

The one square is centered on the worst cluster on the roof. How the desk reads it: if they reinspect and the rest of the slope is clean, every photo in your file is now suspect. Representative squares protect your credibility on the whole claim, well beyond that one slope.

Mistake 3 — Calling weathering "hail"

Blistering, granule loss from age, mechanical damage from a prior repair, manufacturing defects — documented as storm hits. How the desk reads it: an experienced reviewer spots a blister versus a hail bruise instantly, and once they catch one mischaracterization they discount the file. Know the difference cold: hail bruises have a fractured mat you can feel and often a corresponding granule depression; blisters are raised, hollow, and usually pop with a sharp edge of exposed asphalt. Don't claim what you can't defend.

Mistake 4 — No collateral

Roof close-ups only, no elevations, no soft-metal corroboration. How the desk reads it: "Could be old damage, could be foot traffic, no independent corroboration of an event." Dented gutters, bruised AC fins, and torn screens are the easiest, most convincing corroboration you can shoot and they live in Phase 2.

Mistake 5 — Direction-blind photos

Close-ups with no way to tell which slope they came from. How the desk reads it: at best a reinspection request, at worst a quiet denial of the lines they can't place. A slope marker in the test-square frame and slope direction in the filename fixes this for pennies.

Mistake 6 — Stripped metadata and re-saved images

Photos run through three apps, compressed, EXIF stripped, with a visible app watermark. How the desk reads it: weaker evidence, harder to authenticate if the claim escalates. Keep and archive originals; submit clean copies.

Mistake 7 — The destructive cut done first

Lifting and cutting shingles for the layer count before documenting the slope intact. How the desk reads it (if they notice): "Is that storm damage or did the inspector do that?" Document intact, cut last.

Mistake 8 — Overshooting

A hundred and fifty photos, mostly redundant. How the desk reads it: signal buried in noise; the reviewer can't find the four photos that justify each line, so they don't, and they defer. More is not better. Ordered and sufficient is better.

Hail versus wind: documenting different damage signatures

The order is the same but the proof shots differ, and adjusters are trained to look for the specific signature of each peril.

Hail signature

• Random, omnidirectional impacts. Real hail hits the field of the slope in a scattered pattern, not in lines. Your test square should show that randomness.
• Bruising with mat fracture. The defining proof. A circled bruise, a scale object, and a macro showing the cracked mat.
• Granule displacement. Hail knocks granules loose, exposing asphalt; over time those spots oxidize. Fresh strikes show clean asphalt at the impact.
• Soft-metal spatter and dents on gutters, vents, flashing, AC fins — directional and event-dating.
• Directional severity by slope. The storm came from a direction; the slopes facing it should read worse. Document each slope's test square so that pattern shows. A roof with identical damage on every slope regardless of orientation invites suspicion.

Wind signature

• Creasing. A wind-lifted shingle that flops back leaves a crease — a stress fracture line across the tab. Photograph the crease lit from the side so the line is unmistakable.
• Tearing and missing shingles. Document the exposed felt or deck and, if you can find them, the displaced shingles on the ground or in the yard.
• Directional pattern. Wind damage clusters on the windward and leading edges — rakes, ridges, eaves, and the slope that faced the gust. Show the pattern with overviews.
• Seal failure. Lifted tabs that no longer adhere. A gentle lift showing broken seal, photographed without tearing the shingle yourself.

Mixing the two is common in the same storm. Document them separately and label them so the adjuster can attribute lines to the right peril — wind and hail are sometimes treated differently under a policy, and a file that blurs them invites questions.

A worked example: a four-slope hip roof after a hailstorm

Walk through a real sequence so the abstract order becomes concrete. A 2,400-square-foot hip roof, four major slopes, asphalt three-tab, roughly fifteen years of wear, hit by a hailstorm with reported stones around an inch and a half.

Ground (12 photos). Address on the house, mailbox number, street context. Four elevations. On the front elevation, the aluminum gutter shows a row of fresh dents — shot tight. The downspout has spatter marks — shot. The AC condenser on the right side has bruised fins — shot from above so the directional denting reads. Roof overview from the driveway showing all four hips.

North slope (9 photos). Climb to the north slope first because the elevation dents suggested the storm came from the north. Slope overview with the chimney in frame. Chalk a 10x10 square mid-slope where the damage looks typical, shoot it bounded. Walk it: eleven strikes, circle each, shoot it circled. Chalk "N — 11/100" and shoot. Then three proof shots: a circled hit with a quarter, the same hit macro showing fractured mat, and a third hit on a different part of the slope to show the spread.

East slope (9 photos). Same pattern. Overview, square — this one comes up six hits, still well over the typical four-per-square replacement threshold many carriers use, but notably lighter than the north, which is exactly the directional story you want. Six circled, count callout, three proof shots.

South slope (8 photos). Leeward. Overview, square comes up three hits — under the common threshold. You document it honestly anyway: overview, bounded square, circled showing three, callout "S — 3/100," two proof shots. You don't inflate it. The honest light slope actually strengthens the claim, because the directional gradient from 11 in the north to 3 in the south is the unmistakable fingerprint of a real, directional hail event. An adjuster trusts that pattern.

West slope (9 photos). Overview, square at seven, circled, callout, proof shots.

Components (11 photos). Ridge caps cracked from impact — overview and close. Two pipe boots, one with a fresh hail split in the rubber. A turbine vent dented. Drip edge existing condition on two sides. Step flashing at the chimney. Skylight on the south slope, frame dented but glass intact.

Layers and deck (3 photos). Cut at the eave shows a single layer over plank decking — shot of the cut, shot of the plank gap (relevant for a re-deck or new-decking line under code), pitch gauge reading on the slope showing a steep enough pitch to support a steep charge.

Total: about 61 photos, named with numbered prefixes and slope directions, sorted in capture order. The desk reviewer opens the folder and reads, top to bottom: here's the house, here's the storm's collateral on the gutters and AC, here's the directional hail pattern across four slopes with honest counts, here are the components, here are the layers and pitch that justify the labor lines. Every scope line has a photo sitting next to it. There is nothing to hunt for. That file gets approved on the first pass far more often than the same damage shot in random order.

Edge cases that trip up good documentation

Steep and high roofs you can't safely walk

Sometimes the roof is too steep or too high to walk safely, and you should not put a crew member in danger for photos. Drone documentation has matured to where a careful aerial set can carry a claim, but the order still holds: locating shots, elevations, roof overview, then slope-by-slope overviews and zoomed-in passes. The weakness of drone-only sets is the test square and the tactile proof — you can't feel a bruise from forty feet up. Compensate with the highest-resolution close passes your aircraft allows, and be honest with the adjuster about why there's no walked test square. Many carriers now accept drone documentation; some still want a physical inspection for replacement. Know your carrier.

Metal, tile, and wood roofs

The five-zoom model holds, but the damage signatures change. Metal shows dents and coating fracture, not mat bruising — document the dents with raking light and scale, and watch for inflated claims on cosmetic-only denting that carriers often exclude. Tile cracks and shatters; document broken and slipped tiles and the underlayment exposure beneath. Wood shake splits and the impact signature differs from asphalt entirely. Don't apply asphalt hail logic to these systems; document what's actually there.

Older damage and prior claims

If you find old, oxidized hail strikes alongside fresh ones, you have a harder file. Document the fresh damage clearly and don't try to pass old as new — adjusters can date strikes by oxidation and granule recovery, and a file that conflates them loses credibility fast. If there's a prior claim on the roof, expect the adjuster to know, and document only the new event.

Functional versus cosmetic

Some carriers and some policies distinguish damage that affects function from damage that's merely cosmetic, especially on metal. Your job is to document conditions thoroughly and accurately; the coverage call belongs to the carrier. Document everything you find — the mat fracture, the granule loss, the compromised seal — and let the function-versus-cosmetic determination happen where it's supposed to.

The reinspection

If a claim does come back for reinspection, the adjuster will often want to find your test squares. If you chalked them, they may still be visible; mark their locations in your notes so you or the adjuster can return to the same spots. A file that lets a reinspecting adjuster stand exactly where you stood and see exactly what you counted is nearly impossible to overturn.

A pre-climb and on-roof checklist

Print this, laminate it, put it on every crew's tablet.

Before the climb

• Address / house number photo
• Street context photo
• Front, right, rear, left elevations
• Collateral on soft metals: gutters, downspouts, AC fins, screens, garage door
• Roof overview from ground or drone

Per slope (repeat for each)

• Slope overview with a reference object in frame
• Slope direction marked in frame
• Test square bounded (10x10) and shot wide
• Test square hits circled and shot
• Hit count callout in frame
• 3–4 proof shots: circled hit with scale, macro of mat fracture, spread shot

Components (as present)

• Ridge / hip caps
• Valleys
• Drip edge / rake edge existing condition
• Pipe boots / jacks
• Vents, turbines, ridge vent
• Flashing: step, counter, chimney, sidewall
• Skylights
• Gutters / downspouts
• Detached structures

Layers, deck, complexity

• Shingle layers at a cut
• Deck type and condition
• Pitch gauge for steep charges
• Stories / access for difficulty charges

Before leaving the driveway

• Photos sorted in capture order
• Filenames numbered with slope directions
• Originals archived with metadata intact
• Every scope line you plan to write has a photo to match it

Documenting alongside the adjuster on a joint inspection

Many claims get inspected with the adjuster on the roof at the same time as your rep. Done well, the joint inspection is the fastest path to an approval. Done poorly, it turns into an argument on a hot roof while a homeowner watches. The order still saves you.

Walk the adjuster through your sequence rather than scattering them across the roof. Start at the elevations on the ground and point out the soft-metal collateral before you climb — establishing the event up front frames everything they see afterward. On the roof, take them slope by slope: here's the north overview, here's where I'd chalk a square, let's count it together. Counting together is the single most powerful move available, because a hit the adjuster circled with their own chalk is a hit they cannot later deny. Hand them the chalk. Let them mark a square. Photograph the square they marked.

Shoot your own full set even on a joint inspection. Adjusters carry their own cameras and their own documentation standards, and their photos go into the carrier's file, not yours. If the claim is reinspected by a different adjuster or escalates to appraisal, your independent, ordered set is what protects you. Never assume the adjuster's photos will carry the day; they're documenting the carrier's position, you're documenting yours.

Keep the tone collaborative. The desk reviewer behind the field adjuster reads notes, and an adjuster who writes "contractor was professional, damage consistent with reported event, counts verified jointly" moves a claim faster than one who writes "contractor disputed every line." You document conditions and present them clearly; the carrier makes the coverage call. Arguing causation on the roof rarely changes the outcome — clean, ordered evidence does.

Building the documentation habit into your company

The order only pays off if every crew runs it the same way on every roof, including the ones nobody is checking. That's a training and systems problem, not a photography problem.

Standardize the kit. Every inspector carries the same gear: chalk and a lumber crayon for circling, a chalk line or tape for bounding squares, a quarter or a dedicated scale tool, a pitch gauge, a small dry-erase board for slope and count callouts, and a phone or tablet with the documentation app installed and the camera set to keep location data on. A standardized kit produces standardized files.

Use a single app and a single template. When every inspector uses different software, your office spends hours reconciling formats before anything reaches a carrier. Pick one tool, build the photo order into it as a checklist or template, and require it. The app should enforce the sequence so a new hire physically can't skip the test square — the order becomes the path of least resistance instead of a thing to remember.

Audit a sample. Have a manager review a random sample of submitted files each week against the checklist. The first month is ugly; by the third month the crews internalize the order because they know files get checked. The metric that matters is reinspection rate. Track it before and after you standardize, and the drop will pay for the effort many times over.

Tie it back to where the crews go. The cleanest documentation system still wastes crew hours if inspectors are climbing roofs that were never going to total. Pairing pre-climb prioritization — which roofs are aging out, which roofs a storm most likely battered — with a disciplined on-roof photo order is the full loop: send crews to the doors most worth documenting, then document them in the order the desk reads. One without the other leaks money. Both together is a repeatable, defensible claims operation.

The bottom line for your crews

Train the order, not the photos. Any crew member can be taught to shoot a sharp close-up. The skill that separates a documentation set that pays from one that bounces is the discipline to move from the street to the address to the elevation to the slope to the square to the single mark, slope by slope, every time, and to name the result so a desk reviewer never has to hunt.

Do that and you stop losing days to reinspections, you stop leaving supplement lines unproven, and you build files that hold up if a claim ever goes to appraisal. The roof hasn't changed. The damage hasn't changed. The only thing that changed is that you told the story in the order the person paying the claim was trained to read it.

FAQ

What is the very first photo I should take on a roof claim?

The house number or address. It anchors the whole folder to a specific property and is the foundation of chain of evidence. Follow it with a street-context shot before you photograph any damage. These cost a minute and preempt the most common adjuster suspicion: that the photos aren't from this address or this date.

How many photos should a single-family roof claim have?

Usually 35 to 60 for a standard four-slope hail claim. Roughly 4 to 5 locating shots, 6 to 8 orienting shots, about 9 per slope (overview, test square, proof marks), plus 10 to 15 component and cut photos. Under 25 means you're probably thin on slope overviews or test squares; over 100 usually means redundant close-ups that bury the evidence the adjuster needs.

What is a test square and why does it matter so much?

A test square is a marked 10-foot by 10-foot area (100 square feet) in which you count and circle every storm strike. It converts "there's damage" into a hit-per-square number, which is the language carriers use to decide repair versus replacement. Shoot it three ways: bounded and wide, with hits circled, and with the count written in frame. Place it where damage density is typical for the slope, not on the worst cluster.

How do I make sure the adjuster knows which slope a photo came from?

Shoot a slope overview with a reference object (chimney, vent, valley) before any close-ups, put a slope-direction marker in the test-square frame, and include the slope direction in the filename. Direction-blind close-ups are one of the top triggers for reinspection requests and quiet line denials.

How should I name and order the photos so the adjuster can follow them?

Use zero-padded numeric prefixes so files sort in capture order, leave gaps between slopes (10, 20, 30) so you can insert photos later, and put the slope direction and subject in the name, like 11_north_test-square-bounded. If your claims software maps photos to scope lines, name to its convention so the reviewer can match filename to line item without thinking.

How do I document hail versus wind differently?

The order is the same, but the proof shots differ. For hail, show random omnidirectional impacts, bruises with a fractured mat, granule displacement, and soft-metal spatter, plus a directional severity gradient across slopes. For wind, show creasing (a stress fracture across the tab lit from the side), torn and missing shingles with exposed felt, and seal failure. Label them separately, since policies sometimes treat wind and hail differently.

Why do my claims keep coming back as repairs instead of replacements?

Most often because the file is all close-up marks and no map. Adjusters can see the marks fine; what they can't see without slope overviews and test squares is the density and location that justify a full replacement. Add representative test squares on every slope with honest hit counts and a directional pattern, and the replacement case becomes self-evident.

Should I keep the original photos or just the ones I submit?

Keep the originals, untouched, with EXIF metadata intact. Submit clean copies. If a claim escalates to appraisal or litigation, original timestamped and geotagged files are your strongest evidence, while images run through several apps and stripped of metadata are much weaker.

Can drone photos carry a roof claim on their own?

Often yes for documentation, and the same five-zoom order applies: locating shots, elevations, roof overview, then slope overviews and zoomed passes. The weakness is that you can't perform a tactile test square or feel a bruise from the air, so use your highest-resolution close passes and be candid with the adjuster about why there's no walked square. Some carriers still require a physical inspection before approving replacement, so know your carrier's policy.

How does roof-selection or storm modeling help if it can't document damage?

It can't and shouldn't replace your photos. Tools like RoofPredict give a per-roof age range from aerial imagery and storm physics modeled per individual roof, so you prioritize which doors to inspect first instead of climbing every roof in a subdivision. Age is a range, not an install date, and the storm model is odds, not proof of damage. It sharpens your route to roofs worth documenting; the evidence that gets a claim paid is still the test square and the circled hit you shoot on the roof.