Early-stage growth in nature rarely stays exponential for long; something eventually runs out. The logistic model, used here, bends the curve toward a fixed ceiling as it fills up, instead of assuming resources are endless the way the plain population growth calculator does for shorter, earlier horizons.
The logistic equation
P(t) = K / (1 + ((K − P₀) / P₀) × e−rt). Early on the curve looks exponential; as P approaches K the growth slows; the inflection point sits at P = K / 2.
S-curve vs J-curve
An exponential J-curve assumes unlimited resources. The logistic S-curve adds a feedback term that slows growth as the population approaches the natural growth limit, matching almost every real ecological and market system. Early on, before the population reaches even 10% of K, the two curves are nearly indistinguishable: growing from P₀ = 100 at r = 20% for 5 periods, the logistic model gives about 267 while pure exponential growth alone would predict about 272, a gap of only 2%. The feedback term only becomes noticeable once the population is a meaningful fraction of the ceiling.
Applications
Population dynamics, viral epidemic spread (see the viral growth calculator for the early exponential phase of an outbreak), product adoption (Bass diffusion model), tumor growth and yeast cultures all follow the logistic shape. The intrinsic growth rate r controls steepness; K controls the asymptote.
Worked example
Start with P₀ = 100, an intrinsic rate r = 20% per period, and a carrying capacity K = 10,000, the defaults loaded above. After 40 periods, (K − P₀) / P₀ = 99, e^(−0.2 × 40) = e^(−8) ≈ 0.000335, so P(40) = 10,000 / (1 + 99 × 0.000335) = 10,000 / 1.0332 ≈ 9,679, or 96.8% of capacity. The inflection point, where growth is fastest, falls at t = ln(99) / 0.2 ≈ 22.98 periods, when the population crosses P = K / 2 = 5,000.
How the curve accelerates then slows
Tracking the same default inputs across time shows the S-shape directly in the numbers rather than just the chart.
| Periods (t) | P(t) | % of K |
|---|---|---|
| 10 | 695 | 7.0% |
| 20 | 3,555 | 35.5% |
| 30 | 8,030 | 80.3% |
| 40 | 9,679 | 96.8% |
| 50 | 9,955 | 99.6% |
Between t = 10 and t = 20 the population multiplies more than fivefold; between t = 40 and t = 50 it barely adds 3 percentage points of capacity. The same growth rate r produces both a rapid climb and a near-flat plateau, purely because of where the population sits relative to K.
How the intrinsic rate controls speed
Solving P(t) = 0.9K for t shows that the time to reach 90% of capacity equals a constant divided by r: t ≈ 6.79 / r. At the default r = 20%, the population reaches 90% of K at about t = 34 periods. Halving the rate to r = 10% doubles that time to about t = 68 periods, since t is inversely proportional to r. The rate changes how fast the curve unfolds, not its S-shape.
Estimating r and K from real data
Fitting a logistic curve to real observations means estimating both r and K rather than assuming them. A common approach uses the early, still near-exponential portion of the data, well below K, to estimate r the same way the growth rate calculator solves for a rate from two points, then estimates K from where growth has visibly started to plateau. A poor estimate of K, more than of r, is the most common source of a bad long-run forecast. Refitting K periodically as new data arrives, rather than locking it in once, keeps the projection honest as the true ceiling becomes clearer.
FAQ
What is logistic growth?
Logistic growth is exponential growth that slows down as it nears a fixed ceiling called carrying capacity K, producing an S-shaped curve instead of an ever-steepening one. The formula is P(t) = K / (1 + ((K − P₀) / P₀) × e^(−rt)). Starting at P₀ = 100 with K = 10,000 and r = 20% per period, the population reaches about 9,679 after 40 periods, 96.8% of capacity.
What is carrying capacity?
Carrying capacity, K, is the population size an environment can sustain indefinitely given its food, space, and other resources. Growth is positive below K, negative above it, and zero exactly at K. In the default example with K = 10,000, the growth rate falls to zero once the population reaches 10,000, no matter how quickly it was climbing earlier.
What is the difference between logistic and exponential growth?
Exponential growth has no ceiling. It keeps multiplying by the same factor forever. Logistic growth is capped by carrying capacity K, so its rate of increase shrinks as the population nears that limit. A population growing exponentially from 100 would pass 10,000 and keep climbing, while the logistic version levels off near K = 10,000 instead.
When should I use a logistic model?
Use a logistic model whenever growth faces a real limit: animal populations bounded by food or territory, epidemics bounded by the pool of people left to infect, technology adoption bounded by market size, or bacterial cultures bounded by nutrients in a flask. If no such limit exists yet, plain exponential growth is simpler and usually accurate enough.
What is the inflection point?
The inflection point is where the S-curve is steepest, accelerating before it and decelerating after it, and it always sits at P = K / 2. In the default example, K / 2 = 5,000, reached at about t = ln(99) / 0.2 ≈ 22.98 periods, just before growth visibly starts to slow toward the ceiling.
Can logistic models predict the future?
They fit S-shaped historical data well, but the forecast is only as good as the assumed carrying capacity K. Underestimating K makes the model flatten out too early and understate the long-run total, while overestimating K keeps predicting fast growth long after the real system has already slowed.