The plot above shows the total U.S. cases of COVID-19 from The New York Times, based on reports from state and local health agencies^{[1]}.
The solid, upward-sloping line is the number of cases plotted against the logarithmic right axis. This would be a straight line for pure exponential growth. (Why exponential growth?)

The points are estimates of the growth rate as a fraction of the number of cases at any given time. The growth rate for a date is estimated with a log-linear regression over the preceding ten days (11 points if all dates are present). The estimates are significant at better than one percent, meaning there is more than 99 percent confidence the true value is within plus or minus one-third of the estimate.

You can think of the growth rate as a percent growth rate, where 0.10 represents a ten percent increase per day. Early in the pandemic the growth rate was very high, surpassing 31 percent on 24 March 2020. This is partially an artifact of the chaos that occurs early in a pandemic as diagnoses are first differentiating the new virus from other similar diseases. Some of this is also rapid spreading before the public was aware of the disease and its infectiousness. By mid May 2020 the growth rate began settling down to below two percent per day, and has varied between one percent and two percent since.

[1]The New York Times. (2020). Coronavirus (Covid-19) Data in the United States. https://github.com/nytimes/covid-19-data