Pricing needs to be about the future, and in particular, assessing pricing decisions by their impact on future income and future costs. 

The means avoiding two common pitfalls about pricing;

1. Setting prices by allocating overheads and fixed costs on some basis across products and individual policies. These costs are sunk and will be incurred regardless of how much income you generate.   The better approach in to focus only on future variable costs and seek to maximise the contribution from future sales, over whatever timescale aligns with your planning horizon.  By definition this means you must have an understanding of your price elasticity. Then if your chosen prices aren't enough to cover your sunk costs and provide the desired profit margin, you don't have a pricing problem, but a strategic decision about whether you can reconfigure the business, through changes to marketing and operational processes to return the business to profit.
2. Imagining that you can recovery past losses through future price increases. To the extend that past losses reflect in change in claims performance that is expected to continue then that should be reflected in future pricing, but not to recover losses incurred on business already accepted.  In any competitive market, such an attempt will almost certainly fail.  You will be attempting to over-charge future customers. Other players, especially new entrants, who don't have past losses, will be able to grow whilst also achieving good returns at your expense. Many years ago in the days of market tariffs, and what would now be considered as market collusion, such behaviour would have worked as the market all took the same stance. But not in the competitive open UK personal lines markets on the 21st century.