Get my free audit
Get my free audit

Diagnostic

What three seconds costs, in your own numbers.

Four numbers you already know, run against three published speed studies, to produce a rounded monthly range. It is arithmetic on what you type, not a measurement of your site.

The curve

Everything below comes off these two lines.

Two published findings on one time axis. The calculator reads the same five points and does nothing else with them.

Additional conversions available, as a multiple of today's rate 2x 1x 0x 1s 3s 5s 10s Mobile visitors who abandon, upper bound 53% 0% 1s 3s 5s 10s Page load time in seconds
Plotted from the five points the calculator uses. Additional conversions available: none at one second, 0.75 to 1 times today's rate at three seconds, 1.5 to 2 times at five seconds, and no further growth after five. Mobile abandonment upper bound: none up to three seconds, 53 percent immediately above three seconds, flat out to ten. The lower bound of the visitor range is zero at every load time, which is why the shaded area in the lower chart starts at the axis itself.

Your numbers

Slow-Site Cost Calculator

Set the four values, then read the rounded monthly range. Directional math, not a promise: the audit gets specific.

Not sure uses the disclosed 1% to 3% planning range.
Possible monthly visitors lostNot calculated
Possible monthly revenue leftNot calculated
Possible payback for the Presence packageNot calculated

Directional math, not a promise: the audit gets specific.

How the range is worked out
  1. The conversion range linearly interpolates between the 1 second and 5 second Portent observation, then stops growing.
  2. The visitor range uses zero as its lower bound and 53% as its upper bound only above 3 seconds.
  3. The not-sure option uses a 1% to 3% current conversion range. It is an editable planning assumption, not a cited benchmark.

What this measures

Four inputs, three studies, one range.

Your monthly visitors, your current mobile load time, what a new customer is worth to you, and your current conversion rate. The load time picks a point on the curve above. The visitor figure is your traffic against the abandonment upper bound, the revenue figure is your current conversions against the recovery multiple and your customer value, and both are rounded before they are shown.

The payback line divides the first package price by the estimated monthly revenue, which is why it moves whenever the package price moves. If you pick Not sure for conversion rate, it uses a 1% to 3% planning range and says so, because that one number is an assumption rather than a cited benchmark.

Known limits

What this arithmetic cannot know.

  • It has not looked at your site. Every figure comes from the four numbers you typed. Change the customer value and the answer changes with it.
  • The research behind it is aggregate. Portent's ratio came from a large mixed sample; your industry, your traffic mix and your price point can sit a long way from that average.
  • It cannot tell you that speed is your constraint. If people leave because the page never says what you do or what it costs, a faster page loses them sooner.
  • The visitor range has a floor of zero at every load time, on purpose. The honest lower bound for visitors you would have kept is none.
  • Payback is a ratio, not a forecast. It assumes the estimated revenue appears and then continues, which no calculator can establish.

Sources

Where the numbers come from.

Related reading

The 3-Second Problem for Local Business Sites

Where the three second line on the chart comes from, what Google's own research says a slow page costs a local business, and how to tell whether yours is affected.

Read the article

Next step

The audit replaces the guess.

The audit measures your actual pages and says which of these numbers is real for you. It is a person reading the site against what your customers are trying to do.

Request a human audit