Case Study: 10,235 Lots in 6 Weeks at One Auction House

In short: A high-volume estate auction house in the Mid-Atlantic region cataloged 10,235 lots from 38,542 photos across 37 jobs in six weeks using Gavelist. Based on industry labor benchmarks of $3–$5 per lot, that workload represents an estimated $30,705–$51,175 in manual cataloging labor — handled by a single operator instead of a four-to-five-person team.

The Operation

The customer profiled here is a professional estate auctioneer running multiple online sales per week across several bidding platforms. They've used Gavelist as their cataloging engine since March 2026, and the production data covered in this case study runs from March 20 through May 2, 2026 — a six-week window that captures both their typical cadence and their peak-volume jobs.

The numbers from that window:

• 37 completed cataloging jobs
• 38,542 photos processed
• 10,235 lots cataloged
• 3.9 photos analyzed per lot (average)
• 277 lots per job (average)
• 539 lots cataloged per active day across 19 working days
• Largest single job: 868 lots generated from 3,376 photos
• Range: 17 lots on the smallest job, 945 lots on the largest by lot count

This is not a pilot deployment. It's a working production environment where cataloging volume directly determines how many sales the house can close per month.

What They're Selling

Estate work is famously heterogeneous, and this operator's category mix reflects that. Out of 10,235 lots:

Eighteen distinct categories appear in the data. A single estate truck-out can produce Depression glass, vintage Snap-On tools, mid-century lamps, a box of Hot Wheels, and a signed lithograph — and every one of those needs an accurate title, description, and category tag before it can go live. For more on how mixed-lot estates are typically handled, see our guide to AI cataloging for multi-platform auctioneers.

The Manual Baseline

To frame the workload honestly, we anchor to published industry benchmarks rather than to internal speed claims.

According to Auction Item Manager (AIM, 2025), manual lot cataloging — including photo review, title writing, description, condition notes, and category tagging — runs $3 to $5 per lot in fully-loaded labor cost at labor rates of $14–$28/hour and throughput of 15–25 lots per hour. According to Estimint's cataloging analysis, manual cataloging of a 200-lot sale takes 46–64 hours — roughly 14 to 19 minutes per lot.

Apply those numbers to 10,235 lots:

• Labor cost (manual): $30,705 – $51,175
• Working time (manual): 2,388 – 3,241 person-hours (at 14–19 min/lot)
• Headcount to sustain 539 lots/day manually: four to five full-time catalogers, assuming a 7-hour productive day

Cataloging labor is widely cited as one of the primary operational bottlenecks for estate auctioneers — it gates how many sales a house can run per month. We dig deeper into that dynamic in the auction industry technology gap.

The Gavelist Configuration

This operator runs a deliberately lean setup:

1. One operator uploads photos and reviews output. No dedicated cataloging team.
2. Multi-photo grouping — averaging 3.9 photos per lot — gives the AI multiple angles, makers' marks, condition shots, and scale references before it generates a description.
3. Category-aware output spans all 18 of their selling categories with consistent field structure.
4. Direct export to their primary online bidding platforms, no manual reformatting.

The 3.9 photos-per-lot average is worth noting. It's higher than the typical 2–3 photos we see from operators who are still learning the system, and it correlates with better description quality. Our internal recommendation has always been 3 to 5 photos per lot for best results, and this customer's results bear that out.

Reading the Job Distribution

The 37 jobs in this window aren't evenly sized. They cluster into three clear patterns:

• Small consignments (17–100 lots): Typically single-consignor cleanouts or specialty additions to a larger sale.
• Standard estate sales (200–500 lots): The bread-and-butter weekly cadence.
• Full-estate sales (700–945 lots): Multi-day pickups from larger homes.

The largest job — 868 lots cataloged from 3,376 photos in a single push — is the kind of workload that historically forces auction houses to either delay the sale, hire seasonal help, or cut corners on description quality. None of those happened here. For a deeper look at how to plan photography for jobs of this scale, see photographing estate sale items for maximum bids.

What "539 Lots Per Active Day" Actually Means

Across 19 active working days, the operator averaged 539 lots cataloged per day. We're framing this as throughput, not speed — meaning total finished, exportable lots leaving the system per working day, including the operator's review and edit time.

To match that throughput manually at Estimint's benchmarked rate of 14–19 minutes per lot, you'd need:

• At 19 min/lot: 170.7 person-hours/day → well beyond any team's capacity
• At 14 min/lot: 125.8 person-hours/day → roughly 18 full-time catalogers

Even at the fastest manual rate of 15–25 lots per hour (AIM, 2025), sustaining 539 lots/day requires 21.6–35.9 person-hours — about 3 to 5 full-time catalogers.

Either way, what one operator does with Gavelist would require a meaningful team to replicate. We've explored the staffing economics of this shift in the real cost of manual auction cataloging.

Honest Limitations

A few caveats we want on the record:

• Photo quality still governs output quality. Blurry, dim, or mark-obscuring photos produce weaker descriptions regardless of how many you upload. This customer's photography workflow is mature; new operators see lower quality until they tune their setup. See AI-powered auction cataloging.
• High-value specialty lots still need expert review. For signed art, rare coins, fine jewelry, and serial-numbered firearms, this customer flags lots for specialist review before publishing. The AI provides the structural draft; the expert validates attribution and value-relevant detail.
• Category accuracy varies by category. General estate goods, tools, and pottery cluster well. Designer handbags and rare coins are categories where we recommend tighter human review — covered in our AI auction description software guide.
• 6 weeks is a snapshot. It's a representative window for this operator, but seasonality matters. Spring estate volume tends to run higher than mid-summer in the Mid-Atlantic.

What Changed Operationally

When we asked the customer what shifted in their business after deploying Gavelist, three operational changes came up — none of which are about speed in isolation:

1. Sale frequency increased. The cataloging step stopped gating how many sales they could schedule per month.
2. Headcount stayed flat. They didn't need to hire seasonal catalogers for spring estate season, historically their busiest stretch.
3. Description consistency improved across categories. When one operator reviews all output against a consistent template, formatting drift across catalogers disappears. This matters for buyer trust on platforms where description style signals professionalism.

This customer's experience is consistent with the broader pattern: when cataloging stops gating the schedule, sale frequency goes up. According to EstateSales.net (2024), hybrid online and in-person sales saw a 50% increase in sales volume — a trend that rewards operators who can catalog fast enough to keep up with demand.

FAQ

Does Gavelist replace experienced catalogers entirely? No, and we don't recommend treating it that way. It replaces the repetitive structural work — title, description, condition language, category tag, photo grouping — so that experienced staff can focus on attribution, valuation, and exception review. The customer in this case study runs all high-value lots through specialist review before publishing.

What kind of photos does this require? Multiple angles per lot, maker's marks captured clearly, and adequate lighting. This customer averages 3.9 photos per lot, which is in the recommended range. You can catalog with fewer photos, but description quality drops. More guidance is in how to catalog an estate sale.

Will it work for a smaller auction house? The economics scale down. At 277 lots per job — this customer's average — a one-person operation handles a full estate sale per cataloging session. The smallest job in the dataset was 17 lots, which is well within range for a part-time operator.

Bottom Line

10,235 lots. 38,542 photos. 37 jobs. 19 active days. One operator.

The AIM (2025) benchmarks suggest the manual-labor equivalent runs $30,705 to $51,175 and requires a team of four to five full-time catalogers to sustain the daily throughput observed here. The customer ran it with one person and standard estate-grade photography — across 18 categories spanning Depression glass, Snap-On tools, Hot Wheels, and signed lithographs.

We frame this as throughput, not speed. The point isn't how fast any single lot gets cataloged. The point is what one operator can finish, review, and ship to bidders in a working week — and what that does to the business model of an estate auction house.

Sources

• Auction Item Manager (AIM), "Tracking Cost Per Lot: The KPI Every Auctioneer Should Know." aimhq.com
• Estimint, "AI Auction Cataloging for Auction Houses." estimint.com
• EstateSales.net, "2024 Estate Sale Industry Report." estatesales.net
• Gavelist internal production data, March 20 – May 2, 2026 (anonymized customer).