Why do contractor quotes vary so much?

Because two quotes can describe genuinely different jobs while looking alike on the cover. On a deck the swing comes mostly from the hidden work: whether the ledger is flashed and bolted with the fasteners the code names, whether the footings are poured below the frost line or just blocks resting on the dirt, and whether the framing is sized for the snow load your county actually gets. Fold in the guardrail, the stairs, the permit, and old-deck removal, and two straight bids for the very same backyard can land thousands apart without anybody marking up a nickel. The material choice alone accounts for a real chunk: the identical deck costs around $6,800 more built in composite than built in wood, on Zonda's figures.

What is included in a deck estimate?

A finished estimate spells out the decking material and grade, the deck's size and how high it stands, how many footings and how deep, how the ledger attaches and flashes and what fasteners it uses, the framing and its hardware, the guardrail and stairs, and who owns the permit, the inspection, and hauling the old deck off. When several of those are blank, read them as questions to raise before you sign rather than reasons to panic. A line of material and a single total isn't an estimate yet, it's a number.

What is the most expensive part of a deck?

It depends on how you count, and that ambiguity is exactly where bids diverge. By materials, the decking boards themselves are usually the single biggest line, and moving from treated pine up to composite lifts the whole job by something like a third. But by labor, the most expensive part is almost always the part you never see: the footings, the ledger connection, and the framing, which together can be more than half the work on a raised deck and none of which shows up in a photograph. That is why a bid can look cheap on the surface and be missing most of its value underneath. If you want one answer: on a low, simple deck it is the boards, and on a tall one it is the structure holding them up.

Why Is One Deck Quote $8k and Another $22k?

Published by TradeIQ · Updated July 12, 2026

A deck quote is really two prices rolled into one: what you pay for the part you walk on, and what you pay for the part that holds it in the air and pins it to your house. When one bid reads $8,000 and another reads $22,000, that second, hidden price is nearly always where they part ways. So the sharper question isn't why is one deck quote $8,000 and another $22,000 out of plain greed, it's which structural and safety work the cheaper number quietly dropped. Read a deck bid as a list of what made the page and what got left off, and the gap stops looking like a mystery.

A homeowner and a deck contractor review a quote on a tablet beside a backyard deck

Key takeaways

When two deck bids sit thousands apart, the difference is almost always the work you cannot see: the ledger, the footings, and the code guardrail. It is rarely one contractor simply charging less for the very same deck.
A nailed ledger is not a cheaper option, it is a prohibited one. IRC R507.8: attachment "shall not be accomplished by the use of toenails or nails subject to withdrawal." The code also requires hot-dipped or stainless fasteners, flashing, and hold-down tension devices.
You will read that 90 percent of deck collapses start at the ledger, credited to the CPSC. It traces to one engineer writing "I believe" in a 2017 trade column. The CPSC never published it, so we do not repeat it.
Composite explains part of your gap honestly. On an identical deck it costs about $6,800 more than wood, so confirm both bids name the same board before you judge anybody.
In freezing climates, footings must reach below the frost line. Precast blocks resting on dirt are the corner an inspector red-tags fastest on a raised, attached deck.
A deck escapes the model code's permit exemption if it fails any one of four bars: over 200 square feet, over 30 inches above grade, attached to the house, or serving the required exit door. "No permit needed" on an attached deck is false by definition.
Write one deck spec, make every contractor bid that same spec, and read the final numbers only once every other line matches up.

Start with the ledger, where the money and the danger both hide

The biggest thing you can't see on a finished deck is the ledger, the board that fastens the whole structure to your house. It's also the worst place to shave a dollar, and it is the one part of a deck the building code refuses to leave to opinion.

Section R507.8 of the code bans the cheap version outright. A deck attached to a wall "shall be positively anchored to the primary structure and designed for both vertical and lateral loads," and then, in words worth reading twice: "Such attachment shall not be accomplished by the use of toenails or nails subject to withdrawal." A nailed ledger is not a weaker option. It is a prohibited one. The same section closes with a line that tells you what an inspector is supposed to do about it: "Where positive connection to the primary building structure cannot be verified during inspection, decks shall be self-supporting." If nobody can prove the deck is properly tied to your house, the code says the deck is not allowed to lean on it at all.

Around that sit four more requirements a thin bid tends to skip. The ledger has to be a nominal 2x8 at minimum, in treated southern pine, incised treated hem-fir, or approved naturally durable No. 2 lumber. Its fasteners "shall be hot-dipped galvanized or stainless steel," and how far apart they go is set by a table that depends on your joist span and your snow load. The ledger has to be flashed so water cannot reach the band joist behind it. And the deck needs hold-down tension devices to resist it pulling away from the wall: either two of them rated at 1,500 pounds, set within 24 inches of each end of the deck, or four rated at 750 pounds, each option tied to its own detail in the code.

The expensive bid prices all of that: cutting the siding back, flashing the ledger, bolting into solid framing, and adding the ties. The bargain version lags the board straight through the siding and leaves the tension devices off the drawing entirely. It photographs the same on move-in day, then rots and separates a few winters later.

Then the footings, where deck blocks fake what concrete does

Below the frame, the same corner is sitting there to cut. Footings are the concrete piers the deck rides on, and in any climate that freezes they have to reach below the frost line so heaving ground can't shove them upward. Precast blocks that just sit on the dirt are dramatically cheaper, and on a raised, attached deck they are the item an inspector red-tags fastest.

How far down the footings have to reach is a number your building department sets, not a figure off a calculator, since it tracks your local frost depth. So a quote worth trusting states how many footings, how deep, and poured in what. Silence there is simply your first question back. And if one bid says concrete piers and the other says blocks, stop comparing the totals. You are not looking at two prices for one deck. You are looking at two different decks.

The one price gap we can actually put a number on

We used to print a column of dollar deltas here, one figure per line item, and every last one came from a lead-generation site that never published a method. They are gone. The material gap is the one number we can still stand behind, because a single outfit prices both versions of the same deck and shows its work.

Zonda, whose deck figures we lean on over at our cost guide because it is the one outfit that publishes its method, prices the identical deck twice: once in wood and once in composite. Everything else is held constant, the frame, the stairs, the railing, the footings. The composite build lands about $6,800 higher, which is roughly a third again on top of the wood one. So if one of your bids says treated pine and the other says composite, several thousand dollars of your gap is spoken for before anybody has cut a corner. That delta stretches with the square footage, with the board line you pick inside a brand, and hardest of all with the railing, since a matching composite rail costs far more than a wood one. It shrinks on a small deck, and it stops applying altogether if the deck sits low enough to skip a guard, because then you are only paying the premium on the boards. Confirm both bids name the same board and the same grade before you read the totals as a verdict on anyone's honesty.

What the rest of the spread is hiding

Past the ledger, the footings, and the boards, the leftover distance between two bids is a stack of smaller cuts, each easy to leave unwritten. We are not going to invent a price for these, because nobody publishes one we would stand behind. Here is the part that still helps you: each one is a real line, each costs real money, and a bid that never names it has never priced it. Make them price it.

Joist hangers and rated fasteners, rather than toenailed joints. Wherever a joist lands against the face of the ledger or a beam, the code wants an approved hanger carrying it, not a few angled nails.
Corrosion-rated hardware. The preservatives in treated lumber attack ordinary steel, and they do it deep in the joint where no inspection will ever catch it. The code names which metals are acceptable, and it gets stricter still within 300 feet of salt water, where stainless stops being an upgrade and becomes the rule.
Framing sized to its actual span and its actual snow load. Joists and beams cut to the code table versus stretched thin is the line between a firm deck and a trampoline, and it is invisible once the boards are down.
The guardrail. The code stops treating a guard as optional once the deck stands high enough, and the trigger is fussier than the flat 30 inches people quote at each other: the drop is measured at any spot out to 36 inches horizontally from the edge (IRC R312.1.1). Leaving the guard off, or building it light, is one of the largest single savings available to a bid that means to underquote you.
Stairs. A thin bid prices the fewest steps it can and bills the rest later.
The permit and the inspection. If nobody is named to pull it, ask who is.
Tearing out and hauling away the old deck. A line a lowball number likes to skip, and then charge you for.

Set your bids next to each other with that list in hand and the $14,000 you were staring at turns into a column of specific choices, each with a name. The gap was never a mystery. It was itemized, on the more expensive page.

The 90 percent figure, and why it isn't on this page

You'll read everywhere, and you used to read here, that 90 percent of deck collapses start at the ledger, credited to the Consumer Product Safety Commission. We went looking for the study. There isn't one. The number traces to a 2017 trade-magazine column by Frank Woeste, a wood engineer and professor emeritus at Virginia Tech, whose actual sentence was "I believe 90 percent of all deck collapses result from the failure of the deck ledger-to-house connection," based on his own tracking of news reports. The CPSC has never published it. Nobody has a clean count, because the federal injury database has no category for structural failure. What isn't in dispute is where decks fail: two decades of load testing at Virginia Tech is what produced the ledger fastener table the code now prints. The engineers went looking at the ledger for a reason.

Want a second read that isn't trying to win the job?

Send your deck bids through TradeIQ and a vetted builder who isn't quoting the work reads them the way an inspector would, telling you where a ledger spec reads vague or a footing line looks thin, and whether the low total is a real deal or a change order in disguise. Your information stays in-house, and no contractor buys it.

How a lowball deck bid tips its hand

Some warnings aren't in the math at all. They're in how the bid gets sold to you. Read each one as a spot to slow down and ask a direct question, and remember that a bid tripping several of them is previewing how the whole build will go:

A deck that plainly needs a permit, pitched to you as one that somehow does not. The model code exempts a deck only if it clears four bars at once: no more than 200 square feet, no more than 30 inches above grade at any point, not attached to the house, and not serving the required exit door. An attached deck fails that test on the third bar alone, no matter how small it is.
No footing count, depth, or concrete anywhere on the page, or a wave-off like 'blocks are plenty for this.'
A ledger described only as 'attached,' with nothing about flashing, cutting the siding back, or what it bolts into.
'Regular nails and screws' quoted where the code names hot-dipped galvanized or stainless, and nails at the ledger, which the code prohibits outright.
Guardrail, stairs, demo, and disposal blended into one round number with no breakdown behind it.
Cash only, a heavy deposit before a single footing is poured, or a shrug at putting the job in writing.
No contractor license where the state demands one, and nothing to prove liability or workers comp coverage.

How to line two deck bids up so the numbers mean something

The only fair way to read wildly different bids is to force them onto one spec, then let the prices land where they land. Draft a single description of the deck you actually want and make every contractor bid that exact thing. A workable order:

1Pin down the deck itself: square footage, how far it stands above the ground, and the decking material and grade, spelled out so nobody is quoting a different board.
2Write the ledger into the spec: flashed, bolted to solid framing with hot-dipped or stainless fasteners, and carrying the hold-down tension devices the code names. Treat this as the line that cannot be value-engineered away.
3Write the footings in too: how many, how far below your frost line, and poured concrete rather than surface blocks.
4Line up the framing: joist and beam sizes for your snow load, spacing, steel hangers, and corrosion-rated hardware.
5Count the guardrail and the stairs, and pin the hardware grade on each.
6Settle the scope edges: who pulls the permit, who books the inspection, who tears out and hauls the old deck.
7Weigh each warranty as part of the price. A bid with nothing in writing is quietly cheaper for a reason.
8Come to the totals only once the other lines match. By then the expensive bid usually reads as the complete one, and the cheap bid as the deck with pieces missing.

So don't read the low number as either a trap or a trophy. Read it as the bid with the most still unsaid, and make it prove otherwise before it earns your signature. Every dollar of that $14,000 spread has a name once you ask for it, and the contractor who itemized more usually isn't the expensive one. He's the one who wrote down the parts that hold a full deck of people safely off the ground, instead of betting you wouldn't clock that they were gone.

Let a builder with no bid in it read your quotes

Hand over the deck estimates you've gathered. A vetted builder who isn't chasing the job reads them the way they'd inspect their own work, flagging which ledger and footing details hold up, what a low number skipped, and whether the price gap is craftsmanship or missing scope. The builder reading yours has nothing riding on which quote you choose, or on whether you choose any.

A Written Review is the right format for this, and you'll usually have it back inside a day.

Get an unbiased expert review

Frequently asked questions

Why do contractor quotes vary so much?: Because two quotes can describe genuinely different jobs while looking alike on the cover. On a deck the swing comes mostly from the hidden work: whether the ledger is flashed and bolted with the fasteners the code names, whether the footings are poured below the frost line or just blocks resting on the dirt, and whether the framing is sized for the snow load your county actually gets. Fold in the guardrail, the stairs, the permit, and old-deck removal, and two straight bids for the very same backyard can land thousands apart without anybody marking up a nickel. The material choice alone accounts for a real chunk: the identical deck costs around $6,800 more built in composite than built in wood, on Zonda's figures.
What is included in a deck estimate?: A finished estimate spells out the decking material and grade, the deck's size and how high it stands, how many footings and how deep, how the ledger attaches and flashes and what fasteners it uses, the framing and its hardware, the guardrail and stairs, and who owns the permit, the inspection, and hauling the old deck off. When several of those are blank, read them as questions to raise before you sign rather than reasons to panic. A line of material and a single total isn't an estimate yet, it's a number.
What is the most expensive part of a deck?: It depends on how you count, and that ambiguity is exactly where bids diverge. By materials, the decking boards themselves are usually the single biggest line, and moving from treated pine up to composite lifts the whole job by something like a third. But by labor, the most expensive part is almost always the part you never see: the footings, the ledger connection, and the framing, which together can be more than half the work on a raised deck and none of which shows up in a photograph. That is why a bid can look cheap on the surface and be missing most of its value underneath. If you want one answer: on a low, simple deck it is the boards, and on a tall one it is the structure holding them up.

Related Decking resources

How to Tell If a Deck Contractor Is Licensed and Insured How Much Should a Deck Cost Per Square Foot?