How to Make an Educated Guess for Good Forecasting

It is crucial in every company to make some predictions about future of the business. While some forecasts fail, others succeed in providing realistic and correct predictions, which contribute to the success of the company. There are several techniques for making good forecasts, but the bottom line is the same: Forecast fundamentally is an educated guess and information increases the accuracy.

One of the best examples of an educated guess was made by Michael J. Burry regarding the subprime mortgage crisis. He saw it coming years before anybody else did and here’s how he did it: He used the existing information for analysis.

There are five general types of forecasts, however, let’s first drive through these three basic questions.

questions forecasts swift polling


  1. What is the purpose of the forecast and exactly how will it be used? This helps the forecaster to understand how much and what resources will be used in the process.
  2. What are the parts and components of the system or event about which the forecast is going to be made? This helps those involved to have a better understanding of the interacting variables.
  3. How big is the past’s role in forecasting the future? To answer this question, take into account big changes in the system (not recent ones) and critical events of the past.

Now that we singled out the three main questions, let’s talk about the five general methods for forecasting.


1. Naive forecasting

This is the base of forecasting. It is simply looking at past data without considering causal factors, seasonal variations or cyclical trends. Forecasters using this method might just take the data of the same period from the past or use the average from that period.

2. Qualitative and quantitative

Qualitative forecasting is based on personal opinions. Some examples are the Delphi method (The Delphi method is a forecasting method based on the results of questionnaires sent to a panel of experts. Several rounds of questionnaires are sent out, and the anonymous responses are aggregated and shared with the group after each round. The experts are allowed to adjust their answers in subsequent rounds. Since multiple rounds of questions are asked and the panel is told what the group thinks as a whole, the Delphi method seeks to reach the correct response through consensus.), informed opinions and the historical life cycle analogy(A judgmental forecasting technique based on identifying history that is analogous to a present situation, such as the history of a similar object or event, and using that past pattern to predict future). The quantitative method solely consists of analyzing past numerical data. Anything with data in form of numbers could be a source for the quantitative method.

3. Causal forecasting

This method makes an assumption that mathematical function which uses current variables can be also used to forecast the future value of variables. For example, information about ticket sales of a movie could be used to predict the number of movie-related T-shirt sales as well.

4. Judgmental forecasting

Swift Polling, which provides various poll and survey-related services, is the perfect tool to get started. Some of the services are: real-time SMS and web polling, PowerPoint integration, Facebook and Twitter integration, as well as open-ended questions. All those tools can be used to conduct high efficiency forecasting. For example, many betting websites started using polls regarding upcoming games, where users can vote the outcomes of the available matches.. You cannot individually interview all the people with knowledge in a given field, however, you can connect with them through polls. Just keep in mind the targeting. You wouldn’t want nonprofessionals to take part in your survey or poll, in that case you would get an imprecise forecast.   

[inbound_button font_size=”20″ color=”#00e100″ text_color=”#ffffff” icon=”” url=”” width=”” target=”_blank”]CREATE A POLL[/inbound_button]


5. Time series

This type of forecasting employs historical data to predict future outcomes. Data is recorded over a specified period, such as a company’s sales by quarter. This is because past patterns have a tendency to repeat themselves. Time series forecasts are mostly for long-term projects.

The most successful people are ones who can see what’s coming before anybody else does, so use this knowledge to make informed predictions. You and your company can profit from the success.

Like this article?

Share on facebook
Share on twitter
Share on linkedin

Leave a comment

Read Next

Make your next event the best yet
with Swift Polling.