Shark PowerDetect Robot Vacuum Review 2026: A Great Vacuum With a Mop Bolted On

Quick Answer: The Shark PowerDetect robot vacuum is one of the best carpet-and-pet-hair robots you can buy near its street price, and one of the worst moppers. Vacuum Wars measured 88% deep-carpet cleaning and 91.5% pet-hair pickup on the flagship UV Reveal, and gave the vac-only model a 4.52/5 performance score against a 3.56 average — but recorded the mop laying down 3.4–5.3 grams of water per test area against a 0.95–1.1 gram average, producing “one of the lowest combined mopping scores we’ve recorded.” The other thing to know before you shop: “PowerDetect” is not a model. It spans three robot series, three retail-channel prefixes, and a completely separate line of stick vacuums. Buy it for carpet and pet hair on a discount; buy Roborock or eufy if you actually want a mop.

Most robot vacuum reviews end with a hedge. This one does not. The Shark PowerDetect line has an unusually clear split personality: the vacuuming is legitimately excellent, measured against a large field of competitors, and the mopping is the worst-performing part of any major brand’s 2026 lineup by the numbers.

That makes it easy to recommend — as long as you buy it for the right job, and as long as you can work out which of the eleven-odd SKUs wearing the PowerDetect badge you are actually looking at.

Shark PowerDetect by the numbers

• 88% deep-carpet cleaning and 91.5% pet-hair pickup on flattened 2.5-inch carpet, measured by Vacuum Wars on the PowerDetect UV Reveal — both well above its multi-year test averages.
• 4.52/5 performance score for the vac-only PowerDetect at Vacuum Wars, against a 3.56 average, with crevice pickup in the top 10 of over 100 robots it has tested.
• 3.4–5.3 g of water per test area versus a 0.95–1.1 g average — the over-wetting that defines the mop, leaving floors damp for 30+ minutes and, per Vacuum Wars, not adjustable in the app.
• No Pascal rating exists. Shark publishes none for any PowerDetect robot; measured sealed suction on the UV Reveal was 0.95 kPa against a 0.99 average.
• 4.1 inches tall against a 3.8 inch average, per Vacuum Wars — the most common owner complaint.
• 24.5 mm threshold climb on the UV Reveal via NeverStuck, one of the line’s genuine strengths.
• $519.99 to $1,299.99 across the current lineup, with the same SKU listed at wildly different prices by channel.

First: PowerDetect is a name, not a product

This is the part that costs people money. Shark applies “PowerDetect” across three robot series and an entire separate category of stick vacuums, and the model codes do not read like a hierarchy.

Per the Vacuum Wars Shark buyer’s guide, the series numbering runs 3000 = UV Reveal, 2900 = ThermaCharged, 2800 = PowerDetect, with 2700 Stratos, 2600 Matrix Plus and 2500 AI Ultra sitting below as older, non-PowerDetect lines. Only the 2800, 2900 and 3000 series are PowerDetect robots.

Then there are the prefixes. Vacuum Wars spells them out: RV is the retail version sold on Shark’s own site, AV is Amazon-exclusive, and UR goes to wholesale and club channels like Costco. These are the same machine at different prices — an RV2820AE listed at $600 while the AV2820AE ran $519.99.

Two more traps worth naming:

• Shark sells PowerDetect cordless stick vacuums and uprights too (IP3251, IP3256, IP1251). They share the marketing name and nothing else. Reddit’s complaints that PowerDetect auto-adjustment is “a gimmick” are almost entirely about the sticks, not the robots.
• The RV2820AE launched in 2023 as the “Shark Detect Pro Self-Empty” and was rebranded to PowerDetect in September 2024. Consumer Reports still lists it under the old name. Same robot, two names, and any review you find of the “Detect Pro” robot describes this machine.

A pricing note you should not skip: the RV2820ZE was listed at $499.99 by Vacuum Wars and $999.99 on Shark’s own site — the same SKU, at the same time. Shark discounts hard and inconsistently by channel and by week. Treat every figure above as a range, check at least two retailers, and never pay list.

Shark’s robot prices move fast and the docks ship in large boxes, so it pays to have delivery sorted first: get your new robot vacuum in two days — try Amazon Prime free for 30 days.

What the detection features actually do

Unlike most robot-vacuum marketing, PowerDetect describes real, separable behaviours. Shark’s own product pages define them:

• DirtDetect — senses concentrated dirt and boosts suction over it.
• EdgeDetect (dry) — “uses blasts of air to push debris into the robot’s cleaning path.” No competitor does this. It is the single most distinctive thing about a Shark robot, and RTINGS confirmed the edge cleaning “works well on bare floors.”
• EdgeDetect (wet) — the mop pad extends outward to reach edges and corners.
• FloorDetect — identifies floor type and automatically avoids carpet while mopping.
• StainDetect — takes multiple passes over areas it reads as stained.
• NeverStuck — lifts the robot over thresholds and holds the mop pad clear of carpet.

The verdict on whether they work splits cleanly by product category. On the robots, Vacuum Wars found the detection systems genuinely drive the strong performance scores, and RTINGS credited NeverStuck with climbing most obstacles. On the stick vacuums, owners on Reddit describe the auto-adjustment as a gimmick that pauses while it makes up its mind. If you read a scathing PowerDetect take, check which product it is about.

Vacuuming: this is why you buy it

The numbers here are strong and they come from a lab that has tested over a hundred robots.

On the UV Reveal flagship, Vacuum Wars measured 88% deep-carpet cleaning, 91.5% pet-hair pickup on flattened 2.5-inch carpet, 20 of 24 obstacles avoided, and coverage of roughly 2,093 square feet per charge — earning an overall 3.83/5 and a place in its Top 20, with the outlet calling it “Shark’s best robot yet.”

On the cheaper vac-only PowerDetect, the performance score was 4.52/5 against a 3.56 average, with pet hair at 3.85 against 3.42 and crevice pickup in the top 10 of 100+ robots tested. It missed the Top 20 for one reason only: it has no mop.

Two caveats sit against that. First, the 2800-series battery is weak — Vacuum Wars measured a 3,200 mAh cell and scored it 1.30/5 against a 2.17 average, describing it as “only about half the capacity of many competitors.” The UV Reveal fixes this comprehensively, scoring 4.93/5. Second, RTINGS — which has reviewed only the RV2820ZE, published December 2024 — rates pet-hair pickup on carpet as poor, alongside mediocre fine-debris pickup on carpet. That sits in tension with the Vacuum Wars figures, so treat carpet pet-hair performance as model-dependent rather than a line-wide guarantee.

For where this lands against the rest of the field, see our best Shark robot vacuum rankings and the Roborock vs Shark head-to-head.

Mopping: the part that undoes it

Here is the flaw, and it is not marginal.

Vacuum Wars measured the UV Reveal releasing 3.4 grams of water on low and 5.3 grams on medium per evaluation area, against an average of 0.95 to 1.1 grams. That is up to five times the average. Floors stayed wet for 30-plus minutes. Its combined mopping score came out at 0.55 against a 22.5 average — described as “one of the lowest combined mopping scores we’ve recorded.”

The ThermaCharged model is worse. In a direct head-to-head, Vacuum Wars measured it releasing nearly 5 grams of water per evaluation area — the second-highest water output it has ever measured — causing streaking, with no way to turn the water down in the app. Its recommendation was blunt: buy the cheaper standard NeverTouch Pro instead of the ThermaCharged.

Owner reports match the lab. Reddit threads describe dirt streaks left behind after mopping despite frequent pad changes, with at least one owner switching the robot to vacuum-only mode permanently.

This is the single clearest reason to treat PowerDetect as a vacuum with a mop attached rather than a true 2-in-1. If mopping is your priority, our best robot mop and best mopping robot vacuum guides point at machines built the other way round — spinning pads with active water extraction rather than a flat pad and a generous water pump.

Shop PowerDetect mop pads & filters on Amazon →

The 2026 security disclosure you should read first

This is not a recall, and it is not a defect in the vacuum itself — but it is the most consequential Shark robot story of 2026, and no spec sheet mentions it.

In July 2026, security researcher tokay0 disclosed that an overly permissive AWS cloud policy allowed a certificate extracted from one Shark robot vacuum to send root-level commands to other Shark devices in the same AWS region — potentially exposing stored home maps, plaintext Wi-Fi passwords and, on camera-equipped models, live camera feeds. The disclosure was reported by The Hacker News, Tom’s Hardware, Malwarebytes and Vacuum Wars, among others.

Three qualifiers matter for keeping this in proportion. The initial certificate extraction required physical access to one device; everything after that was cloud-side. Camera exposure applies only to camera-equipped models, which most Shark robots are not. And no outlet has independently reproduced the exploit — the reporting all traces to a single researcher’s writeup.

SharkNinja acknowledged the report in March 2026 but had issued no fix, public statement or CVE as of July 20, 2026. The fix is server-side — a rescoped cloud policy in SharkNinja’s own account — so no firmware update is required from owners, which also means there is nothing you can install to resolve it. The only user-side mitigations are disabling remote control or keeping the robot off Wi-Fi.

Separately, and worth stating plainly because it often gets conflated: there is no CPSC recall of any Shark robot vacuum, PowerDetect or otherwise.

Specs at a glance

Measured figures are from Vacuum Wars; RTINGS data is from its RV2820ZE review of December 2024. Modern Castle has not tested a PowerDetect robot, and Consumer Reports scores for the RV2820AE and RV2920XE sit behind its paywall — so where this table says “not published,” that is the honest state of the public record rather than an omission.

Known weaknesses

1. Over-wetting when mopping. Up to five times the average water output, not adjustable in the app, 30+ minutes of drying, and streaks. Worst on ThermaCharged.
2. Weak battery on the 2800 series. 3,200 mAh, roughly 100 minutes, scored 1.30/5 by Vacuum Wars. Fixed on the UV Reveal.
3. Obstacle avoidance. RTINGS scored hazard recognition 0.0 on the RV2820ZE and called it ineffective. The flagship improves to 20 of 24, but the mid-range models sat at 12 to 16.
4. App limitations. RTINGS calls the app “limited in overall functionality”; Vacuum Wars notes the UV Reveal lacks mop zones and multi-level mapping.
5. 4.1 inches tall. Above the 3.8-inch average and the most frequent owner complaint — sofas and toe kicks it cannot reach.
6. High recurring maintenance cost, flagged directly by RTINGS.

How it compares

At the $500–$800 street prices where PowerDetect actually sells:

• vs Roborock — Roborock’s spinning-pad mopping with active water extraction beats Shark’s decisively, and its retractable-LiDAR flagships are nearly an inch shorter. Shark’s counters are EdgeDetect air-blast edge cleaning, which nobody else offers, and NeverStuck climbing. See Roborock vs Shark.
• vs Roomba — Tom’s Guide has noted that iRobot is no longer the default best choice in 2026, with Shark and eufy winning head-to-heads at lower prices. Our Shark vs Roomba comparison covers the detail, and the Roomba Max 705 review shows what iRobot’s answer looks like at nearly twice the money.
• vs eufy — eufy is the direct value rival, and it wins exactly where Shark loses: mopping and obstacle avoidance. Shark vs eufy is the matchup to read.

Who should buy the Shark PowerDetect

• Buy it if your home is mostly carpet and your problem is pet hair — the measured carpet and pet-hair numbers are genuinely top-tier for the money.
• Buy it if edge and corner debris is your frustration; the EdgeDetect air blast has no equivalent anywhere else.
• Buy it if you have thresholds and transitions that stop other robots — 24.5 mm of NeverStuck climb is real.
• Buy the discounted 2800-series NeverTouch Pro rather than the ThermaCharged; Vacuum Wars recommends the cheaper one outright.
• Skip it if mopping matters. This is the clearest verdict on this page.
• Skip it if you need clearance under low furniture — 4.1 inches rules out a lot.
• Skip it if you were counting on the 2800-series battery to cover a large home in one pass.
• Wait if the unresolved cloud-security disclosure concerns you, or plan to run the robot without a Wi-Fi connection.

The bottom line

The Shark PowerDetect is a better vacuum than its price suggests and a worse mop than its box suggests. Vacuum Wars measured 88% deep-carpet cleaning and 91.5% pet-hair pickup on the UV Reveal and a 4.52/5 performance score on the vac-only model against a 3.56 average — that is flagship-grade cleaning at mid-range money, backed by an edge-cleaning trick nobody else has. Then the same lab measured a mop putting down up to five times the average amount of water with no way to turn it down, and called the result one of the lowest mopping scores it has ever recorded.

So buy on the strength, not the badge. If you want a carpet-and-pet-hair machine and you shop the $519.99 to $749.99 discounts rather than the $999.99 and $1,299.99 list prices, the PowerDetect is one of the better buys of 2026 — just check the four digits after the RV, AV or UR prefix so you get the NeverTouch Pro base, and read the July 2026 security disclosure before you connect it to your network. If mopping is why you are shopping at all, spend the same money on a Roborock or a eufy instead. Start with our best robot vacuum rankings if you are still weighing the field.

Check today’s Shark PowerDetect price on Amazon →