This area of the Land Registry continues to confuse quite a lot of people. I have explained it before in other posts, but it seems like it needs a dedicated discussion.

If you have ever looked at UK house price data and thought “that doesn’t feel right”, there is a decent chance the issue is not the market.

It’s Category A vs Category B sales.

These two labels appear innocently enough in Land Registry data, yet they are routinely misunderstood, mislabelled, and occasionally blamed for price movements they had nothing to do with.

So let’s clear it up properly, because this distinction matters far more than most dashboards let on.

The Short Version (For People Who Will Scroll Anyway)

• Category A = standard open-market sales
• Category B = non-market or discounted transactions
• Neither category tells you whether a home is new or old

That last point is where things usually go wrong.

What Category A Sales Actually Are

Category A sales are normal market transactions.

That means:

• A willing buyer
• A willing seller
• No special relationship between them, or at least it does not impact the normal market value
• No discount for being related, employed, or favoured

In practice, Category A includes:

• Existing homes sold via estate agents
• New-build homes sold by developers at full market price
• Most mortgage-backed purchases

In other words, this is the dataset people usually think they are looking at when discussing house prices.

Because of this, Category A sales are what feed:

• The House Price Index (HPI)
• Average price statistics
• Year-on-year comparisons
• Most “the market is doing X” commentary

If you want to know what buyers are genuinely paying in competitive conditions, this is the category that matters.

What Category B Sales Actually Are

Category B sales are still legitimate transactions, but not open-market ones.

They involve a transfer of ownership and a recorded price, but that price may not reflect market value.

Common examples include:

• Right to Buy purchases
• Sales between family members
• Transfers between companies or related parties
• Housing association disposals
• Discounted employee or tenant sales
• Certain repossessions or special-term sales
• Parking spaces can be sold separately from a property, but still have a title number, so they appear as a sale

The key feature is not the property; it is the terms of the sale.

A perfectly normal three-bed house can be Category B if it changes hands at a discount or under special conditions. Equally, a brand-new house can be Category A if it is sold openly at full price.

The Mistake People Keep Making

At some point, someone decided that:

“Category B = new builds”

This is wrong.

Persistently wrong.

And quietly destructive to analysis.

New builds are usually Category A.

They are sold openly, marketed widely, and priced competitively, exactly what Category A is designed to capture.

Category B exists to remove distorted prices from market statistics, not to label construction types.

Why Category B Exists at All

Imagine mixing these Identical property sales into the same dataset:

• A £400,000 open-market purchase
• A £220,000 Right to Buy discount sale
• A £1 transfer between companies

You would still technically have “prices”, but they would tell you absolutely nothing useful about market behaviour.

Category B is there to say:

“This sale happened, but please don’t use it to infer price trends.”

And for good reason.

Why This Matters More Than People Think

When Category A and B sales are mixed:

• Average prices fall for reasons unrelated to demand
• Certain areas appear “cheap” for entirely administrative reasons
• Apparent price crashes appear where none exist

This tends to happen:

• In areas with lots of Right to Buy
• Around large housing association stock
• In datasets scraped without filtering

The result is confident commentary built on quietly broken numbers.

How I Handle This on PropertyResearch

I separate Category A and Category B explicitly.

• Market analysis is based on Category A
• Category B is visible, but treated as what it is: non-market data
• Charts and summaries do not quietly mix the two

This avoids accidental pessimism, false optimism, and the sort of charts that look dramatic but explain nothing.

The One Line to Remember

Category A and Category B describe how a property is sold, not what the property is.

If you take nothing else from this post, take that.

Final Thought

If a dataset looks strange, it usually isn’t lying; it’s just being misunderstood.

And Category A vs Category B is one of those details that seems small, until you realise it quietly underpins almost every property chart you’ve ever seen.