Over the past few weeks, I’ve been taking a look at various analytical techniques that may be appropriate for understanding more about visitor behavior than you may find in your average Web analytics tool. Many of these techniques, like classification and segmentation, involve the use of statistical analysis tools. This week, I’ll continue in that vein by looking at forecasting and some of the techniques used to assess and understand future trends.
As businesses build their data trends, those trends become more interesting and useful. One problem with a fast-growing environment in which all charts show lines shooting up and to the right is it’s difficult to know what the underlying trends are and whether marketing activity is affecting this growth.
There are two broad categories of forecasting techniques: quantitative methods and qualitative methods. Quantitative methods are based on algorithms of varying complexity, while qualitative methods are based on educated guessing. I’ll focus on quantitative methods here. In part two, I’ll look at qualitative methods.
Quantitative methods come in two main types: time-series methods and explanatory methods. Time-series methods make forecasts based purely on historical patterns in the data. Say you want to forecast site visitors over the next few weeks. Time-series methods only use historical site visit data to make that forecast.
Explanatory methods use other data as inputs into the forecasting data. In the previous example, you might include marketing data as inputs into a model to understand how they affect visit levels and to forecast future visits with those data. These types of techniques have been used for ages in the offline world to evaluate marketing activity’s effect on brand awareness or sales.
Time-series methods are probably the simplest methods to deploy and can be quite accurate, particularly over the short term. Most quantitative forecasting methods try to explain patterns in historical data as a means of using those patterns to forecast future patterns.
Simple time-series methods include moving average models. In this case, the forecast is the average of the last “x” number of observations, where “x” is some suitable number. If you’re forecasting monthly sales data, you might use a 12-month moving average, where the forecast for the next month is the average over the past year.
Trouble is, simple averaging methods don’t tend to work well when there’s either a trend in the data or seasonal effects. This tends to be the case in a lot of marketing data! In that case, other techniques, such as exponential smoothing, may be more appropriate.
With moving averages, every data point carries equal weight in making the forecast. With smoothing methods, more importance is placed on the most recent data than on the historical data. If there’s a trend in the data, it’ll use the recent observations to make up the bulk of the forecast, and the forecast is more likely to reflect the trend.
Moving averages and simple exponential smoothing techniques are available in Excel and easy to execute. That’s part of the great advantage of time-series methods: they’re generally simple, cheap to run, and relatively easy to interpret.
There are more complex time-series techniques as well, such as ARIMA (define) and Box-Jenkins (define) models. These are heavier duty statistical routines that can cope with data with trends and the seasonality in them. You’d probably need to invest in a statistical analysis package or a dedicated forecasting package to use these more powerful algorithms. Like any analytical technique, though, you shouldn’t use them blindly or treat results as gospel. All forecasts are invariably wrong, in fact. It’s just a question of how wrong they are.
So why would you use these heavier duty forecasting techniques?
Forecasting techniques are often used as much for their explanatory power as for their predictive power. Understanding the trends and seasonal behavior of your business provides a better understanding of its underlying health.
In consumer goods marketing, for example, these types of forecasting techniques are often used to assess a brand’s baseline performance. A forecast is made of what the sales would have been in the absence of certain types of promotions or advertising so underlying trends can be assessed.
Explanatory forecasting methods take the process a step further and allow you to relate changes in marketing activity to changes in such outputs as sales, brand awareness, and registrations. Here, we’re looking for causality and can feed that into forecasts as a way of evaluating marketing response. We’ll take a look at this in more detail next time.
Marketers create personas to better understand their target audience and what it looks like. If marketers can understand potential buyer behaviors, and where they spend their time online, then content can be targeted more effectively.
What’s behind a successful data-driven marketing strategy?
One of the major challenges in the martech industry is getting the attention of prospects in a world where they are bombarded by content and emails on all sides.
Facebook is addressing one of the biggest missing pieces of its chatbot offering: analytics.