If your target is 12% annualised and the forecast volatility of the instrument is 24%, you hold half a unit. When forecast volatility drops to 8%, you hold 1.5 units. The notional moves around a lot; the risk contribution does not.
The benefits are real but specific: more stable drawdowns, a smoother equity-curve, and improved compounded return through reduced volatility drag. It does not raise expected return and it will not save you from a gap, because forecast volatility is always backward-looking.
Costs to budget for: the rebalancing generates turnover and therefore slippage, and the method mechanically increases size in calm markets, which is exactly where leverage-driven blowups originate. Cap the maximum position regardless of what the formula says.
Related: realised-volatility, position-sizing, garch, turnover