ReserveIQ

What is ReserveIQ?

ReserveIQ uses AI to suggest a reserve price for your property based on its data — one that's competitive enough to attract bidders and give the auction the best chance of a successful sale.

In auctions, the instinct to set a high reserve to "protect" a property's value can quietly work against the seller. A reserve pitched a little below expectations often does the opposite of what you'd expect — it widens the buyer pool, draws more attention, and sparks the competitive bidding that frequently pushes the final hammer price above what a cautious reserve would have achieved. ReserveIQ surfaces that sweet spot — grounded in how comparable properties have actually performed at auction — so you can set a reserve with confidence that it's working for the sale rather than against it.

How ReserveIQ Works

ReserveIQ is powered by a machine learning model trained on a large dataset of historical UK property auctions. The model learns the relationships between a property's attributes — its type, size, location, condition, and more — and the prices those properties actually achieved at auction. It then uses those patterns to estimate a reserve price likely to result in a successful sale for your property.

Importantly, this is not a generative AI or Large Language Model (LLM). LLMs are excellent at writing but poorly suited to numerical prediction — they tend to hallucinate or default to round, popular numbers rather than analyse the underlying data. ReserveIQ instead uses a specialised machine learning regression model that produces a consistent, data-grounded estimate every time it's run for the same inputs.

The model is retrained periodically as more auction data becomes available, so its accuracy improves over time.

Reading the Chart

The chart sits in the Pricing tab, alongside your reserve price input. As you enter or adjust the reserve, the pointer moves along the chart to show how that figure compares to the model's prediction.

The chart is split into five colour bands. From right to left:

• Good (green) — your reserve is at or below the predicted price.
• OK (light green) — your reserve is slightly above the prediction.
• High (yellow) — your reserve sits noticeably above the prediction.
• Too high (orange) — your reserve is well above the prediction.
• Way over (red) — your reserve is far above what the model expects.

How to Use It

1. Fill in the Property Details

The prediction is generated automatically once the property record has enough information. Make sure you have entered:

• Postcode (on the Location tab)
• Property type (on the Property tab)
• Bedrooms count (on the Property tab)

Until those are populated, the panel will prompt you to add them.

2. Set your Reserve Price

Navigate to the Pricing tab. If your branch has an AuctionPro subscription, you'll see the ReserveIQ panel on the right-hand side of the form. Type your reserve into the Reserve field — the chart updates a moment after you stop typing, and the pointer position and colour show at a glance how your reserve compares to the model's prediction.

If you later change the postcode, property type, or bedroom count, the prediction itself will refresh automatically.

What Data Does It Use?

The prediction is based on the following property attributes:

• Property type
• Bedrooms, bathrooms, and reception rooms
• Square footage
• Postcode
• City
• Latitude and longitude
• Auction type
• EPC rating
• Council tax band
• Property condition

The model is trained on historical auction data from comparable residential properties to suggest a reserve that's competitive in the local market.