What Really Determines Airbnb Prices in London

We studied 85,432 active listings across London to understand pricing dynamics. The findings challenge common assumptions about revenue optimization. While hosts invest considerable effort in response times and acceptance rates, these factors explain less than 1% of pricing variation.

Instead, location and property type dominate the pricing equation. A property in Westminster commands a £42 premium over identical properties in outer boroughs. An entire flat generates £107 more per night than a private room. These are structural advantages that operational excellence cannot replicate.

The implications are significant for hosts, investors, and policymakers. For hosts, this means pricing strategy should anchor on property fundamentals rather than behavioral optimization. For investors, it validates location-first acquisition criteria. For policymakers, it quantifies the affordability pressure created by short-term rental concentration in central boroughs.

Central London boroughs command a consistent £42 per night premium over other areas. This differential persists even when controlling for room type, host response metrics, and other variables. The statistical evidence is robust: a t-statistic of 54.14 (p<0.001) with a tight 95% confidence interval of [£40.68, £43.74].

What makes this finding particularly striking is its independence from other factors. We isolated the location effect by examining properties of the same type with similar host characteristics. The premium remains consistent. Geography is destiny in London's short-term rental market.

Room type is not a minor variable in the pricing equation. It represents fundamental product segmentation. Entire homes average £209 per night, private rooms £94, and shared rooms £54. These are not marginal differences requiring careful statistical analysis to detect. They are obvious, large, and economically meaningful.

The effect sizes confirm this intuition. Cohen's d values exceeding 1.5 indicate very large practical significance. In statistical terms, room type explains more pricing variance than any other single factor we measured. The product matters more than the provider.

This finding has immediate practical applications. Investors considering property conversion should model the trade-offs carefully. A three-bedroom property converted to three private room listings generates approximately £282 nightly (3 × £94), compared to £209 as a single entire-home rental. However, the room-by-room approach requires more operational complexity, guest management, and potentially different licensing.

This is where our findings diverge most sharply from conventional wisdom. Airbnb's platform architecture emphasizes host performance: response rates, acceptance rates, Superhost status. The implicit message is clear: optimize these metrics to charge more.

Our regression analysis tells a different story. Improving acceptance rates from 0% to 100% correlates with just £24 in additional pricing power. Not £24 per night, but £24 total. Each extra minimum night requirement adds 6 pence to the nightly rate. Host response rates fail to reach statistical significance in the multivariate model (p=0.233).