Seeq Is an Application Specifically Built to Handle Time-Series Data
Microsoft Excel is an excellent tool for analyzing data, it’s the industry standard whatever process manufacturing vertical you’re in. But about 90% of the total time process engineers spend in data investigations and report writing is just to get the data ready. By “ready,” I mean selecting, copying, pasting, cleansing, aligning, contextualizing (adding in data from other sources), filling in data gaps, and formatting time series data.
Once the data is ready then spreadsheet applications like Excel shine as fantastic tools for doing aggregation and reporting. But why do engineers also do the first 90% of the work in Excel, toiling for hours-and-hours data wrangling? Because for many years Excel has been the best option to use with process historian data, and thus every process historian vendor ships an Excel export feature to get data from their historian into Excel. But Seeq changes that: Seeq is an application specifically built to handle time-series data, and Seeq’s goal is to give the time you spend wrangling data back to you so you can focus on the important part of your job – actually analyzing, investigating, and finding insights in the data.
To illustrate how Seeq can save time and accelerate time to analysis, I performed a usability test with Seeq in a real-world process and compared the time it took to accomplish the same results with time-series data in Excel and Seeq. The end point was a view for the user to begin analysis – something similar to the view shown in Figure 1 below:

Figure 1: Data ready for analysis
This usability test was done in collaboration with a chemical engineer at the University of Texas – she used Seeq and I acted as a monitor, noting each step she performed and the time it took. Her desired output was getting a set of relevant signals displayed during a specific time period of interest. In Figure 1 the time period of interest is denoted by the pink horizontal bar at the top called a “capsule” in Seeq (for more on capsules see the earlier blog post). In this case the capsule represents a operational run for one of her clients. In the past the chemical engineer would manually locate these times by perusing through many columns and rows of data by eye. With Seeq, she used the signal values to define a capsule that can be visually identified much faster.
The times it took to perform the tasks in Excel were estimated based on her experience doing them many times over generating reports. The results are shown in Table 1 below. The the data formatting task is described along with the time it took to perform them in both Excel and Seeq for a side-by-side comparison.
# | Task | Excel (min) |
Seeq (min) |
Notes |
1 |
Export the signals from the historian to individual Excel spreadsheets. |
10 |
0 | Seeq connects live to the historian - no need to export raw data. |
2 |
Append all the data files into a single spreadsheet with signals shown side-by-side. | 15 | 0 |
A custom Excel Macro was used here - without it this number would have been much greater. Seeq automatically does this "data gridding". |
3 |
Find the time period of interest. | 50 | 10 |
User must manually "eyeball" the right time period in Excel. Seeq creates a capsule set based on a signal value. |
4 |
Add all ~20 signals on a single chart during area of interest. | 200 | 10 |
In Excel you have to do a lot of cut/paste into region of interest for all the variables. With Seeq you can easily add variables and spread out in the display window. |
Totals | 275 | 20 | 93% Reduction |
Table 1: Seeq vs. Excel time comparison
As you can see, by using the only basic features of Seeq: bring in the data, find a time period of interest, and display it on the screen – the user saves a significant amount of time. I witnessed my chemical engineer colleague reduce hers by 93% - that is over 4 hours per report!
Interestingly, knowing what needs to be done in Excel is not the difficult part of the process. In this use case the chemical engineer was adept at exporting process data to the spreadsheet, lining up timestamps, removing invalid entries, etc. She even optimized these steps with the use of custom macros, pivot tables, and proficient drag-and-drops. But doing all of these steps, in varying order, with different source data, within a single spreadsheet was a time-drain. Not applicable in this case, but in some cases it is not even possible to use Excel if the number of records exceeds 1,048,576 rows.
Data wrangling in Excel is where the pain sets in. Users must deftly go through data formatting acrobatics to get the data in an adequate state before analysis can even start. And this takes a lot of time out of the workday – as was the case with our chemical engineer, where more than 90% of the time is spent simply getting the data ready for investigation. What users need is an application that is built exclusively to handle time-series data and quickly enable experts to apply their expertise to extract valuable, actionable insight.
Seeq aims to give you time back otherwise spent formatting and wrangling data in Excel. The goal is to give you more time doing the fun part and applying your expertise to analyze your data. Ironically, with Seeq perhaps you will find yourself using Excel even more – but to analyze data and report your insights.