Modeling Time-to-Sale with Hazard Rates and Survival Curves

One of the most persistent challenges in domain name investing is understanding how long it will take for a particular name to sell. While it is tempting to assume that each year carries a fixed probability of sale, the reality is far more nuanced. The timing of sales matters because it dictates cash flow, influences renewal strategies, and determines whether capital is being tied up productively or stagnating in illiquid assets. To capture this complexity, investors can borrow concepts from actuarial science and reliability engineering, specifically hazard rates and survival curves. These mathematical tools, often used to model lifespans of machines or mortality rates of populations, are surprisingly well-suited to domains because they quantify the chance that an event—in this case, a sale—occurs at a given point in time, and how that chance evolves as the asset ages.

The hazard rate represents the conditional probability of a domain selling in a specific year given that it has not sold yet. For example, if the annual sell-through rate of a certain class of names is assumed to be one percent, then the hazard rate is 0.01 in year one. But the critical insight is that the hazard rate may not remain constant. A newly registered trend domain might have a higher chance of selling in its first or second year when interest in the trend is peaking, and that chance may decline over time as the trend cools. Conversely, a premium generic keyword might actually see its hazard rate rise over time as more businesses are founded and demand for the exact-match name accumulates. Treating the hazard rate as a fixed constant oversimplifies the dynamics of how domains find buyers in practice.

The survival curve complements this by describing the probability that a domain remains unsold over time. In statistical terms, the survival function is one minus the cumulative distribution function of sales. If a domain has a one percent hazard rate that is constant across years, then after the first year, the survival probability is ninety-nine percent. After the second year, the probability of still holding the name is approximately ninety-eight percent, and after ten years it is about ninety percent. This curve provides a visual and mathematical way to understand how inventory decays, not by being dropped but by being converted into sales. For investors with portfolios of thousands of names, survival curves allow forecasting of how many domains will remain unsold after a decade and how many sales are likely to have occurred in that period.

What makes these tools particularly valuable is their ability to incorporate changing hazard rates. Suppose an investor is holding domains tied to a new technology keyword like blockchain. Early on, hazard rates might be three percent per year as businesses scramble for branding, but after five years, interest might diminish, dropping hazard rates to half a percent annually. By modeling a time-varying hazard rate, the survival curve reflects the burst of early sales followed by a long tail of low probability outcomes. This mirrors the real-world phenomenon where certain categories of domains experience hype cycles with high turnover early and diminished demand later. Understanding this pattern helps investors decide whether to renew aggressively in the first few years and then cull names as hazard rates decline.

On the opposite side of the spectrum, consider evergreen names such as strong one-word generics or common service keywords in popular extensions. These domains may have initially low hazard rates because buyers are few and deals take time to materialize. But as businesses mature, funding becomes available, and competitors realize the branding value of the asset, the hazard rate may increase steadily. The survival curve in this case flattens early on but then accelerates downward as more sales accumulate later. This pattern justifies patient holding, even across decades, because the expected payoff grows with time and the hazard rate improves with age. Sophisticated investors factor this into their models, budgeting for long-term renewals knowing that the survival function will eventually converge to significant sales probabilities.

Mathematically, these models can be built using exponential, Weibull, or log-logistic distributions, each of which captures different hazard rate behaviors. The exponential model assumes a constant hazard rate, which is useful for average portfolio-wide estimates but often too simplistic. The Weibull model allows hazard rates to increase or decrease with time, making it useful for both trend domains that decay and evergreen names that mature. The log-logistic model captures situations where hazard rates rise to a peak and then decline, reflecting categories that become hot for a period and then fade. By fitting historical sales data to these distributions, investors can estimate the most realistic hazard functions for their portfolios.

In practice, constructing these models requires historical sales records of comparable domains. By observing how long after registration or acquisition domains in similar categories have sold, one can approximate hazard rates empirically. For example, if sales data shows that two percent of brandable .coms sell within their first two years and only half a percent per year thereafter, then a two-stage hazard model can be applied. The corresponding survival curve will reveal the proportion of names likely to sell in the short-term burst versus those that linger for longer. This process of calibration turns abstract probability theory into actionable forecasts, guiding renewal budgeting and pricing strategy.

The utility of survival analysis extends further when applied to portfolio-level cash flow planning. By combining hazard rates with average expected sale prices, investors can project not just the number of sales in each year but also the revenue distribution. If hazard rates are front-loaded, cash inflows will be concentrated early, allowing faster reinvestment. If hazard rates grow with time, revenues will come later, requiring patience and capital reserves to sustain renewals. The shape of the survival curve becomes a proxy for liquidity planning, dictating how much capital an investor should expect to tie up and for how long before realizing returns.

There is also a psychological benefit to framing sales in terms of hazard rates and survival curves. Many investors become discouraged when names sit unsold for years, interpreting inactivity as failure. But a survival model clarifies that unsold inventory is not necessarily underperforming, it is simply following the statistical expectation. A domain with a one percent hazard rate per year should not be expected to sell quickly, and in fact, survival probabilities suggest that the majority of such domains will remain unsold after a decade. Knowing this allows investors to set realistic expectations and avoid premature liquidation of names that still carry long-term value.

Finally, hazard modeling sharpens decision-making about pruning portfolios. If a category of names shows declining hazard rates over time with survival curves flattening near one hundred percent, that signals weak prospects for future sales and supports dropping names to reduce renewal costs. Conversely, categories with rising hazard rates justify retention even after years of holding. This transforms portfolio management from a reactive, emotion-driven process into one grounded in probabilistic reasoning.

In conclusion, hazard rates and survival curves provide domain investors with a mathematical framework to model the elusive question of time-to-sale. By moving beyond simplistic constant probabilities and instead capturing the dynamics of changing sale likelihoods, these tools allow for more accurate forecasting, better renewal strategies, and more disciplined portfolio management. Domain names may appear static, but their probability of sale is dynamic, shifting with trends, industries, and market cycles. Survival analysis transforms this uncertainty into structured expectations, giving investors the clarity needed to navigate the long game of domain investing.

One of the most persistent challenges in domain name investing is understanding how long it will take for a particular name to sell. While it is tempting to assume that each year carries a fixed probability of sale, the reality is far more nuanced. The timing of sales matters because it dictates cash flow, influences…