Outsourced product data entry: the real cost per SKU

Published rate cards put outsourced product data entry at roughly $0.20 to $0.40 per SKU for simple offshore transcription, $0.33 to $0.75 for standard listings, and $0.80 to $2.00 for complex technical SKUs, with nearshore and U.S. teams charging two to six times more. Behind those prices sit offshore hourly rates of about $4 to $8, so the per-SKU number mostly comes down to how many SKUs an operator can finish in an hour, and that depends on how technical your products are.

That is the number on the quote. The number on your P&L also includes the spec you have to write, the QA you have to run, the rework you have to fund, and the second project you buy when the attribute model changes.

Outsourced product data entry cost per SKU, by complexity tier

The clearest published breakdown comes from Mercury Minds, which ties per-SKU prices to throughput and region:

Tier	Throughput	Offshore	Nearshore	U.S. / specialized
Simple: transcribe existing attributes	15 to 20 SKUs/hr	$0.20 to $0.40	$0.60 to $1.20	$1.00 to $2.33
Standard: full attributes plus a written description	8 to 12 SKUs/hr	$0.33 to $0.75	$1.00 to $2.25	$1.67 to $4.38
Complex: detailed specs, edited images, multiple languages	3 to 5 SKUs/hr	$0.80 to $2.00	$2.40 to $6.00	$4.00 to $11.67

The same source puts 2026 hourly rates at $4 to $6 offshore, $12 to $18 nearshore and $20 to $35 for U.S. or specialized teams. The table is hourly rate over throughput.

Other vendors price by the record or by the seat. Precise BPO Solution publishes $0.01 to $0.05 per record for defined batches, $4 to $8 an hour for flexible work, and dedicated operators from $960 a month. Intellect Outsource advertises data services from $4.00 an hour and a full-time employee at $680 for 170 hours a month.

A "record" is not a SKU. A per-record price usually means one row or one field set in a defined batch, which is why it looks an order of magnitude cheaper than per-SKU pricing for a product that needs forty attributes pulled from a spec sheet. Ask every vendor to quote the same 50 sample SKUs, not a rate.

Product data entry outsourcing hourly rates vs. a U.S. data entry keyer

The domestic benchmark is the BLS occupation code 43-9021, data entry keyers. O*NET, using BLS data, lists a 2025 median wage of $19.88 an hour and about 131,800 workers, with employment projected to decline slightly through 2035. The last full BLS percentile table we could retrieve, May 2023, showed a mean of $19.29 an hour, a 10th percentile of $28,250 a year and a 90th of $55,330. Check CareerOneStop's Salary Finder, which runs on the May 2025 BLS estimates, for your metro.

Zedtreeo lists $4 to $6 an hour for a dedicated offshore specialist against $15 to $25 for a U.S. in-house employee and claims a 70 to 85 percent cost reduction. On wages alone, that math holds. For illustration, a U.S. keyer at the $19.88 median doing complex SKUs at 4 an hour costs about $4.97 per SKU before benefits; an offshore operator at $6 an hour at the same pace costs $1.50.

The catch is "at the same pace," plus everything the rate excludes.

The costs a rate card leaves out

Writing the spec. An offshore team types what the spec tells it to type. Someone on your side has to define, per category, which attributes matter, the allowed values, the units (in or mm, PSI or bar), which source wins when the catalog and the spec sheet disagree, and what to do when a value is missing. For a distributor with 300 leaf categories, that is a real project for a category manager, and it is unpaid on the vendor's invoice.

QA sampling. Vendors publish high accuracy figures. Precise BPO lists 99.6 percent for SKU and attribute entry and 98.5 percent for vendor catalog normalization. You still have to verify it, because accuracy on a sample you didn't pick tells you little about your hardest categories. Pavago recommends spot-checking every 500 records and acting immediately if accuracy drops below 98 percent. Zedtreeo notes that double-key verification pushes cost to roughly 1.8 times single entry.

Rework. Pavago makes the point that fixing a 5 percent error rate across 5,000 records can cost more than paying double upfront. Errors in product data travel: a wrong voltage or thread_size turns into a return, a marketplace suppression or a support call, and the person who catches it is rarely the person who made it.

Management time and ramp. Zedtreeo cites one to two weeks of onboarding; Pavago says two to four weeks for a dedicated offshore hire. Add daily questions across time zones, a weekly accuracy review, and the internal owner who answers "is Stainless 304 the same as SS304?" for the hundredth time.

Turnaround. Precise BPO quotes first records within 48 hours of scope confirmation. The real bottleneck is the round trip on ambiguous SKUs, which waits on your team.

Doing it again. Product data entry is a project; product data is not. When a marketplace adds a required field, when you add a facet like ingress_rating, or when a manufacturer updates 2,000 spec sheets, the outsourced work does not update itself. You re-scope, re-spec and pay again. We cover how this compounds on low-velocity SKUs in the economics of long-tail SKUs.

A worked example: 20,000 technical SKUs

For illustration, take a 20,000-SKU industrial catalog in the complex tier, quoted offshore at $6 an hour and 4 SKUs an hour.

• Vendor labor: 20,000 / 4 = 5,000 hours × $6 = $30,000, or $1.50 per SKU.
• Spec writing: assume 300 categories at 2 internal hours each = 600 hours. At a loaded $60 an hour, $36,000.
• QA: assume a 5 percent sample re-checked internally at 10 minutes per SKU = about 167 hours, or roughly $10,000.
• Rework: assume 3 percent of SKUs (600) need a second pass at vendor cost plus 15 minutes of internal triage each = about 150 internal hours (about $9,000) plus roughly $900 of vendor time.
• Year two: assume 25 percent of the catalog needs updating after spec and channel changes. That is another 5,000 SKUs through the same pipeline.

Under these assumptions the "$1.50 per SKU" project lands at roughly $4.30 per SKU in year one once internal time is counted, and it does not stay done. The hidden cost scales with how technical the catalog is, not with the hourly rate.

Is offshore product data entry cost effective?

Offshore is genuinely the right call when:

• The source data is already structured and the job is transcription, such as keying a supplier price file or moving a clean spreadsheet into a new platform.
• The attribute model is stable and won't change for a year.
• The work is one-time, like a platform migration with a fixed end date.
• You have an internal owner with time to write specs and audit samples.

It stops being cheap when:

• Values live in PDFs, spec sheets, and manufacturer sites, and every SKU needs judgement about which source is right.
• Your categories are technical enough that a wrong value causes a return or a compliance problem.
• Requirements change on a cycle you don't control: marketplace templates, retailer portals, new facets.
• You measure success by attributes filled, and nobody is checking whether they are correct. We break that trap down in fill rate vs. accuracy.

Alternatives to outsourcing product data entry to an offshore team

The three realistic options are an in-house team, an offshore or BPO team, and automated enrichment that reads source documents directly. A fourth, generic AI writing tools, generates descriptions but does not extract or verify attribute values, so it doesn't replace the attribute work. Our offshore data entry vs. automated enrichment guide walks through the decision by catalog type, and the data entry outsourcing comparison lines up the models side by side.

This is where the "who does the work" question gets answered. Your PIM stores the data. Anglera does the work: it extracts attribute values from the actual spec sheets, catalogs, images and manufacturer pages, normalizes units and values to your schema, quality-scores each value, and flags conflicts for review instead of guessing. It works alongside whatever PIM, ERP or syndication platform you run, or from a flat CSV export, and a typical implementation takes 30 days or less. Because it is a maintained practice rather than a batch project, adding a new attribute means backfilling it across the catalog instead of re-scoping a vendor engagement.

How to decide

Price the same 50 hard SKUs three ways, and count your own hours, not only the invoice. If the offshore quote still wins once spec writing, QA and year-two updates are on the sheet, it is the right call. If it doesn't, see how Anglera works on a sample of your own catalog.