Statistical PERT is free so you will try using it

Project managers and product owners need better ways of estimating so they can quickly align the expectations of their stakeholders. Product owners and agile teams need simple, effective ways to forecast when new features and capabilities will be available, while project managers estimate project duration and cost.  Executives need better estimates so they can make better decisions today to secure better outcomes tomorrow.  Everyone needs an easier way to communicate their sense of confidence and risk about future uncertainties.

Statistical PERT® is a simple, powerful way to do all of that.

Everything about this site is intended to break down barriers to a better way of estimating.  That’s why all SPERT® spreadsheets and templates are licensed using the GNU General Public License (GPL), so you can freely download, use, modify and share everything on this site.  That’s why Statistical PERT uses Excel’s built-in statistical functions (so you don’t have to buy anything extra).  That’s why this site doesn’t require you to register before downloading, and why this site uses a secured, encrypted connection.

So how do you make money off this?

I don’t make any money off this. Statistical PERT is my gift to project managers, product owners, agile teams, and anyone else who needs to create probabilistic estimates. Statistical PERT is for anyone who needs to quickly align expectations and make better decisions today to secure better outcomes tomorrow. You’re welcome. 🙂 If you choose to watch my Pluralsight course, Easily Estimate Projects and Products, that will help defray the cost of running this website. I also accept Paypal donations, too. But you’re under no obligation to watch any video or make a donation. Statistical PERT is freely licensed under the GNU General Public License.

If you’re looking for a catch, you can stop looking. There is no catch.

Can I use Statistical PERT with Google Sheets?

No. Well, actually you can, but Google Sheets doesn’t have many of the features or capabilities of Excel, so if you try opening a SPERT spreadsheet in Google Sheets, it won’t look or function very well.  The actual statistical functions that Excel uses are available in Google Sheets (and other spreadsheet programs), but the charting capabilities are not as robust as Excel.  Therefore, Statistical PERT is only supported for Microsoft Excel.

That said, contact me if you’d like to have a Google Sheets-compatible SPERT spreadsheet.  If I get enough interest in a Google Sheets version, I might be inclined to create one.

Can I use Statistical PERT with Excel 2010 or Excel 2013?

Yes and No. With Microsoft’s release of Excel 2021 in October, 2021, I have decided to stop supporting Excel 2010 and Excel 2013 for use with Statistical PERT®. That means some newer features of Statistical PERT spreadsheets (or other spreadsheet models available on this site) may only partly work or may not work at all with Excel 2010 or Excel 2013.

For instance, Microsoft introduced the new histogram chart in Excel 2016, and some Statistical PERT spreadsheets now have a worksheet that uses a histogram chart to represent simulated data. In older versions of Excel before Excel 2016, this chart can’t be rendered, so the simulation results can’t be visualized.

Also, I have used the CONCATENATE function in Statistical PERT spreadsheets, but Microsoft has deprecated the use of CONCATENATE and now supports just the CONCAT function. In later releases of Statistical PERT, I will switch the use of the CONCATENATE function to CONCAT to ensure my spreadsheets remain supported by the latest Excel functions and features.

But as of right now, most features of a Statistical PERT spreadsheet will work just fine provided you are using Excel 2010 or later. Any version of Excel before Excel 2010 will not work with any Statistical PERT spreadsheet.

Statistical PERT uses the normal because it’s good enough

PERT uses a special form of the beta distribution but Statistical PERT (SPERT®) Normal Edition uses the normal distribution because, for many estimation problems, using the normal distribution is good enough to make a good decision.

As long as the bell-shaped curve is only slightly or moderately skewed, Statistical PERT Normal Edition’s use of Excel’s normal distribution functions will still yield results that are comparable to a Monte Carlo simulation of the same uncertainty modeled using the beta distribution.

The key benefit of using the normal distribution is that Excel’s two normal functions, NORM.DIST and NORM.INV, are both useful and very easy to use.  PERT provides an estimate for the mean argument, and Statistical PERT creates the required standard deviation needed by those two Excel functions.

The goal of any estimation effort should be to develop an estimate that is good enough to make a good decision, and that is accurate enough to be within the estimator’s tolerance for error.  While using the beta distribution may be more accurate than using the normal distribution for modeling asymmetrical uncertainties, that fact alone is not a good reason to discard using the normal distribution.  For many estimation problems—even involving asymmetrical uncertainties—using the normal distribution leads to a “good enough” estimate that is within the estimator’s tolerance for error.

That said, on March 1, 2017, I released another edition of Statistical PERT that uses Excel’s two beta functions, BETA.DIST and BETA.INV.  The Statistical PERT® Beta Edition will accurately model skewed probabilities using the beta distribution’s two shape arguments, alpha and beta.

Also, on January 1, 2022, I released the Statistical PERT® Lognormal Edition, which uses Excel’s lognormal functions, LOGNORM.DIST  and LOGNORM.INV, to model uncertainty where the tendency is for an uncertainty’s true outcome to be greater than the “most likely” outcome instead of less than the “most likely” outcome (in other words, the uncertainty is skewed to the right on a bell-shaped curve).

With these two newer editions of Statistical PERT (Beta Edition and Lognormal Edition), anyone who is concerned with modeling uncertainty with right-side skewing can do so without using the normal probability distribution found in Statistical PERT Normal Edition.

Statistical PERT Normal Edition uses a different formula than PERT does to find a standard deviation.  Why?

Statistical PERT Normal Edition replaces the PERT formula with a different formula that lets the estimator use subjective judgment, intuition, emotion, and/or private knowledge to rationally adjust the standard deviation.  The SPERT standard deviation formula (in the Normal Edition) is:  (Max – Min) * RSM, where RSM is the Ratio Scale Modifier that corresponds to the estimator’s subjective judgment about how likely is the most likely outcome.  The SPERT SD formula better conforms the implied distribution curve to a specific uncertainty.

Watch a Pluralsight course to learn more

You can learn more about Statistical PERT® by doing any of the following: