Replace One or Both? A Guide to Moog Ball Joints, Sway Bar Links, and Wheel Bearings

No universal answer—there are four scenarios

I manage procurement for a 14-person independent repair shop. I've handled our parts budget—roughly $340,000 annually—for the past six years, negotiated with 20-plus suppliers, and logged every order in our cost tracking spreadsheet. In that time, I've learned that the "replace one or replace both" question almost never has a clean answer.

Whether you're pricing out a Moog 6696 ball joint, a pair of Moog front sway bar links, a 5/8 wheel bearing, or answering the customer at your counter who just asked "if one spark plug is bad, should I replace all"—the decision usually lands in one of four scenarios. Each one runs on different logic. Mixing them up is how you end up with a comeback job that eats your margin.

Here's how I sort them:

1. Symmetric wear on a high-mileage vehicle. One side failed; the other side is probably on borrowed time.
2. Small-order friction. You need one or two units and a supplier's minimum order policy is in the way.
3. Brand-tier selection. The spec sheets look identical, but the price doesn't.
4. Preventive set replacement. Only one part failed, but the rest of the set may be close behind.

Let me walk through what I actually do in each.

Scenario A: one ball joint or sway bar link failed on a high-mileage vehicle

This is the most common one. A customer comes in with a clunk over bumps, and the diagnosis lands on a worn ball joint—say, the Moog 6696 application they found online—or a worn Moog front sway bar link on one side.

The textbook advice is "replace both sides." But that advice ignores inspection. If the opposite side has 70,000 miles on it and the boot is cracked, yes, replace both. If the opposite side was replaced 8,000 miles ago and still has a firm joint with an intact boot, you're probably fine replacing just the failed side (assuming the customer isn't planning a 3,000-mile road trip next month).

I learned this the expensive way. Back in 2021, I had a customer with a 2010 sedan—one Moog front sway bar link had snapped. I talked him into replacing both "for safety"—or rather, I assumed symmetric wear without checking the service records and priced the pair accordingly. The second link turned out to have been replaced by another shop nine months earlier. He wasn't happy about paying for a part he didn't need, and he told me so (fairly, honestly). I should add that this wasn't a safety issue—I'd just assumed symmetric wear without checking service history.

Now I check three things before recommending a pair replacement: mileage on the opposite side, boot condition, and whether the opposite side appears in the vehicle's service history. If all three point to "probably original and probably tired," replace both. If not, price out the single unit and let the customer decide with real information.

Cost math, for reference: a single Moog 6696 ball joint runs roughly $40–70 depending on supplier and application (based on quotes I collected in Q4 2024; verify current pricing). Labor is the bigger variable—if both sides are accessible in the same bay, the second one adds maybe 40% of the first one's labor. So a "both sides" job isn't double the cost. It's usually 1.4x. That's the number I put in front of customers.

Scenario B: small orders for 5/8 wheel bearings and control arm bolts

Here's where I'll take a position: small orders deserve the same service as large ones. If you're a one-bay shop ordering a single 5/8 wheel bearing and a pack of control arm bolts because that's what the job needs, you shouldn't have to apologize for it.

That's not how every supplier sees it. I've had distributors quote me a "minimum order" of $500 when I needed three control arm bolts—total, about $18. I've also had suppliers quietly wave the minimum because they figured a small order today might turn into a standing account later.

The ones who waved it? I still buy from them. Six years later. Oh, and I've referred three other shops to them. My current bearing supplier started as a $200 order in 2019 (one 5/8 wheel bearing and two seals, if I remember right). Last year we spent about $22,000 with them. They treated my 200-dollar order like it mattered, so I kept coming back.

If you're the small buyer in this scenario, ask directly: "Is there a way to get this order through without the minimum?" Some suppliers have a will-call or stocking-fee option that's cheaper than padding an order with parts you don't need. Some will combine your order with a larger shipment. The worst they can say is no—and the ones who only say "order more or don't order" have told you something useful about how they'll treat you later.

(Which, honestly, is better to learn on an $18 bolt order than on a $4,000 brake job.)

Scenario C: brand selection when the spec sheets look identical

The third scenario is often mistaken for a simple price comparison. It's not. Two ball joints can share the same dimensions and still differ in materials, heat treatment, boot quality, and testing. That's true across the aftermarket, not just for Moog.

What I actually do: compare the specification sheet line by line against the OEM service manual for the vehicle. Moog's chassis line publishes material and tolerance data that you can check against OEM specs—that's the comparison that matters, not "which brand is better" in the abstract. When a supplier can't or won't produce a spec sheet, that's your answer.

The oversimplification trap here is thinking "OE part number equals identical part." An aftermarket part can meet or miss the OE spec depending on which spec you're checking. A Moog 6696 ball joint and a budget unit might both fit, but the boot material and joint preload are where the differences live—and those are exactly the things that determine whether the part lasts 40,000 miles or 140,000.

Practically: I don't claim one brand is universally better. I claim that a supplier who shows you their spec data and stands behind a warranty is a better risk than one who only competes on price. That's a procurement position, not a brand loyalty position—and it's the reason we stock Moog for chassis work but source other categories elsewhere.

Scenario D: the spark plug question—as a model for set replacement

Now the one that started me thinking about all of this: "If one spark plug is bad, should I replace all?"

Customers ask this every week. And the answer is a clean example of Scenario D logic: on a standard four-cylinder engine, if one plug has failed at its service interval and the other three are at the same mileage, replace the set. Misfire balance, ignition coil load, and the fact that plugs are cheap relative to the labor of opening the engine up again all push toward the set. That's not overselling—it's TCO logic.

But—and here's the part the "always replace the set" advice ignores—if the plug failed early because of a coil issue or a wrong heat range, the "replace all" answer might be covering up the real problem. Replacing the set doesn't fix a coil. And if the vehicle just had plugs done 5,000 miles ago, replacing all four again is throwing money at a diagnosis problem, not a wear problem.

The same logic maps onto suspension work surprisingly well. A failed ball joint on a 150,000-mile vehicle is probably a wear failure—check the counterpart. A failed ball joint on a 30,000-mile vehicle might be a defect, an installation issue, or impact damage—replacing the counterpart isn't the answer; figuring out why it failed is. The set-replacement decision should follow the wear diagnosis, not replace it.

How to figure out which scenario you're in

Run through these, in order:

1. Is the failure wear-pattern or event-driven? Wear on one side at high mileage usually means the counterpart is close behind. Impact, defect, or installation failure usually means the counterpart is fine—and the real question is why the part failed.
2. What's the labor overlap? If both sides are accessible in the same bay, pricing the pair is usually only 30–40% more labor than one side. That changes the math. If the second side means a separate appointment and a separate lift, the pair-replacement case gets weaker.
3. How small is the order? If you're buying one or two units, ask about minimums directly before you assume the order is blocked. A surprising number of suppliers can accommodate; the ones who can't have told you something.
4. What does the spec sheet say? If you can't compare the spec against the OEM service manual, you're buying a brand name, not a known quantity.

None of this produces a magical "always do X" rule. That's the point. The right answer depends on which scenario you're actually in—and the cost of mixing them up, in my experience, shows up about six months later as a comeback job or a customer complaint.

(I've eaten both. The comeback from assuming symmetric wear was $260 in labor I didn't bill. The complaint from over-selling a pair replacement cost me a customer review. Neither is a mistake I've made twice.)