A paver surface almost never fails all at once. It gives you a rocking unit by the mailbox, then a shallow dip near the garage door, then a border course that has quietly walked an inch into the grass. Most people live with all three for a year or two before they call anybody.
The good news is that this is the failure mode pavers are built for. Everything below can be opened up, corrected, and closed back with the same stones. What matters is reading the pattern correctly first, because the same dip has three or four different causes and only one of them is the one on your property.
The failure chain, from the top
Under your pavers there is a thin layer of sharp bedding sand, and under that a compacted aggregate base, and under that whatever the lot was made of. Pavers move when one of those layers loses material or loses shape. In South Florida the usual culprit is water going somewhere nobody planned for.
Washed-out bedding sand is the most common single cause we find. Concentrated runoff moves that sand out through the joints and out at the edges, and once enough of it has left from under a small area, the pavers above it have nothing left to sit on. They do not crack the way a slab would. They simply follow the void down.
Base erosion at the edges is the next one. The perimeter of any paver surface is the weakest part because it is the only place the base has an open face. Water running along that face carries the fine material out of the aggregate, the base loses its packed structure, and the outer courses drop and tilt outward. This is why so much failure shows up as a sagging edge rather than a hole in the middle.
Roots are constant here. A paver base holds moisture and stays cooler than open ground, which makes it attractive to root systems from ficus, black olive, oak, and most of the mature landscaping people plant near driveways and walkways. A root under a paver lifts rather than sinks, and it usually shows up as a ridge or a single high corner that catches a toe.
Utility trenches are the one people never guess. If a line was run across your driveway for water, sewer, irrigation, cable, or a pool, that trench was dug and then backfilled. If it was backfilled loose and not compacted in layers, it keeps consolidating for years afterward. The tell is unmistakable once you know it: a straight dip running across the surface in a line, rather than a rounded low spot.
Reading the pattern before touching anything
The shape of the problem tells you most of what you need to know, and you can read it yourself before we ever get there.
A broad dish in the middle of the field, a shallow bowl a few feet across with the pavers all still tight to one another, usually means the base under that area consolidated or was never compacted properly to begin with. The surface followed it down as one piece. That is a base story and it will keep going if only the surface gets touched.
A dropped or rolling edge, where the outer two or three courses have sunk and tipped toward the yard, points at the perimeter. Either the edge restraint has failed or the base has eroded out from under the open face. Look for a gap opening between the border and the field.
A single rocker in the middle of an otherwise solid surface is usually a local void: bedding sand gone from under that one unit, often near a joint where water concentrates. Step on one corner and it tips, step on the other and it tips back. This is the smallest version of the problem and by far the cheapest to catch early.
A straight-line depression is the trench, as above. A ridge or a lifted corner near a tree is the root. And a surface that is generally uneven everywhere, with no single pattern, usually means the original base was thin or the compaction was skipped across the whole job, which is a different and larger conversation.
The other thing worth doing is looking after it rains. Stand outside during a real Florida downpour, or go out right after one, and watch where water actually runs. People are often surprised. The dip is almost always downhill of something.
Why we lift and relay instead of pumping or mortaring
People ask about mud-jacking, or about running mortar under the low spots, because both work on other surfaces. Neither belongs on pavers, and the reason is the whole point of the system.
An interlocking paver surface is designed to be flexible. The units carry load by locking against each other through sand-filled joints, and the whole field is meant to move slightly and settle as a unit over ground that shifts. Injecting rigid material underneath creates a hard spot in a flexible system. The surface then hinges at the boundary between the rigid patch and everything around it, and the new failure line shows up along that seam. You also have not addressed why the void formed, so the water that hollowed it out the first time is still doing its thing around the edges of your new concrete.
Mortaring individual pavers down is worse in a smaller way. Now those units cannot be lifted again without breaking them, and you have given up the one property that made the surface repairable.
So the correct repair is lift and relay. We pull the affected units by hand and set them aside in order. We dig out whatever is left of the compromised bedding and base, and we find out what happened while we are down there, because that hole is where the answer is. Then the base gets rebuilt in compacted layers, screeded fresh bedding sand goes back at the right depth, and the same pavers go back down in the same pattern. Joints get refilled and the area gets compacted back into the field.
Done properly the repair disappears. That is the real argument for owning pavers rather than a poured slab: a slab that cracks and settles has to be broken out and replaced, while a paver surface that settles gets opened, corrected, and closed with the material you already own. Our loose paver repair and base repair work is mostly this, over and over, on surfaces that end up looking untouched.
Edge restraint and the creeping border
Here is the tell almost nobody recognizes for what it is. Look at where your driveway or walkway meets the grass or the planting bed. Is the border course still in a clean line, or has it bowed outward, with the joints between border units opened into visible gaps?
That is edge restraint failure, and it is one of the most common structural problems we find. Every paver surface needs something holding the perimeter in place, whether that is a plastic or metal restraint spiked into the base, a concrete haunch, or a mortared soldier course. Its whole job is resisting the outward pressure the field puts on its own edges every time something drives or walks across it.
When that restraint fails, and around here it fails from spikes corroding, from the base under it eroding, or from being installed on top of soil instead of on the compacted base, the perimeter starts moving outward a fraction at a time. As the border creeps, the joints in the field open. Open joints lose sand faster. Less sand means less interlock, which means more movement, which opens the joints further.
That loop is why a surface that was fine for eight years can seem to go downhill quickly in the ninth. And it is why we look at the edges first even when the complaint is about a spot in the middle. Fixing a dip while leaving a failed restraint in place means being back in two years.
The repair that comes back, and the one that does not
If water hollowed out that base once, it will hollow out the rebuilt one too unless something upstream changes. This is the fork in the conversation that decides whether you are paying us once or every couple of years, and it is the part we insist on having before any pavers come up.
What it means practically is that a relay is sometimes the small half of the work. If the water arriving at that spot has an obvious source we can redirect, the surface fix holds. If the water is arriving because the surface itself sheds the wrong way, or because there is nowhere for it to go once it gets there, then correcting the drainage is the actual repair and relaying the pavers is the finish work that follows it. Sometimes that means resetting a section at a corrected slope while it is open, which is far cheaper to do during a repair than as its own job later.
The reason we bring this up before quoting anything is simple enough. We would rather tell you on day one that your dip has a fixable cause than relay the section, take your money, and let you discover the truth during the next wet season.
Matching old pavers without pretending
When units break during a lift, or when some were already cracked, we have to put something back. Matching honestly is a craft and it involves some tradeoffs worth understanding.
Concrete pavers are made in production runs, and color varies between runs even within the same product line. A unit off today’s pallet in the same name and shape will usually be close but not identical to one made nine years ago. On top of that, your existing surface has weathered, taken sun, and been walked on. A brand new unit next to it reads as brand new no matter how good the color match on paper is.
So the first move is always to use what is already yours. If a paver is sound but the top face is chipped or stained, it often flips. The underside has never seen sun and frequently comes up cleaner than the face. Second, we harvest: pull good units from somewhere nobody looks, under a planter, behind a gate, along the side yard, and use those in the visible repair, putting the new units in the hidden spot instead. That single move is the difference between an invisible repair and an obvious patch.
Where new material has to be visible, we blend rather than block. Scattering new units through a wider area with existing ones between them lets the eye read variation instead of a rectangle of fresh color. It is also worth knowing that a new unit and an old one usually converge over a couple of years of sun, so a repair that reads slightly light at first tends to settle in.
What we will not do is tell you it will be a perfect match when it will not. If your line was discontinued and the closest available unit is visibly different, you deserve to hear that before the work starts, along with the options: a deliberate contrasting band, replacing a whole defined section, or living with a slight difference. There are examples of both repair work and full surfaces in our gallery.
When repair stops being the right answer
Most surfaces are worth fixing. Some are not, and being straight about that line is part of the job.
The clearest case is a base that failed everywhere. If the whole surface is uneven, joints are open across the entire field, and there is no single pattern to point at, the base under it was probably never built for what it carries. Repairing one area at a time on that surface is paying repeatedly for a problem that is going to keep surfacing somewhere new. Rebuilding it once is less money over five years than four repairs.
The second case is material condition. When a large share of the units are spalling, cracked through, or worn down past their color layer, you are relaying stones that are near the end regardless of how good the base under them is.
The third is arithmetic. Once a repair covers a big enough share of a driveway, the cost of pulling, rebuilding, and relaying starts approaching what a proper new surface costs, and the new surface comes with a base built correctly. We will tell you when you are near that line, and we will tell you when you are not, which is more often. Plenty of the calls we take about a driveway repair turn out to be a half day of work and a downspout extension.
Questions we get about sinking and loose pavers
Q. Can I just pull the paver up and put sand under it myself? For one rocking unit in a walkway, honestly, yes, and plenty of people do. Pull it, add sharp bedding sand, tamp it, set it back level with its neighbors, and sweep joint sand in. What that will not do is answer why the sand left. If it sinks again within a season, or if there are several units involved, something below is moving and it is worth having someone look at it properly.
Q. Why do my pavers keep sinking in the same spot? Because the cause was never addressed. A repeat failure in one location almost always means water is still being delivered there, usually by a downspout, an irrigation head, or a grade sending runoff across that corner. Occasionally it is a utility trench still consolidating. The repair keeps failing because it is treating the symptom, and the fix is to change where the water goes.
Q. Will you have to replace all my pavers to fix one sunken area? Almost never. The whole point of an interlocking surface is that individual units come up and go back down. A typical repair involves lifting the affected area plus a small margin around it, rebuilding what is underneath, and relaying your own pavers. New material is usually only needed for units that broke or were already cracked.
Q. My border pavers have shifted outward and there are gaps. Is that serious? It is worth attending to sooner rather than later. Spreading borders mean the edge restraint has failed, and once the perimeter is loose the field joints open and lose sand, which lets everything move more. It is a much smaller job caught at the gap stage than after the field has started to spread and dip.
Q. How long does a repair like this take? Most residential repairs are one day, and many are less. A single rocking unit or a small dip is a few hours. A larger area with base rebuilding runs a full day, and if drainage work is part of it, plan on two. We tell you the realistic answer when we look at it, including the part about needing the area kept clear of vehicles while the base sets up.
If something on your property is sinking, rocking, or drifting out of line, the useful next step is having somebody stand on it and tell you which of these it is. Call 786-238-0499 and we will come see what it is doing, or send us a few photos with a note about where the water goes when it rains. Half the time we can tell you what you are dealing with before we ever pull a paver.