10 Construct Tricks You’ll Wish You Knew SoonerReading time: 6 minutes
I recently hosted a three-part Construct Bootcamp Deep Dive series featuring beginner, intermediate, and advanced training. After one of the sessions, Jeffrey Collis (the most recent addition to our analyst support team) mentioned that he gained a few Construct tricks that would have made his prior work in higher education analysis a lot easier, had he known them earlier.
As the Product Manager for Rapid Insight, many of the tips and tricks I use to speed up my work are second-nature. I barely notice that I’m using them because they’re an ingrained part of my workflow, but Jeffrey’s comment made me realize that many users may not be aware these tricks exist.
With that in mind, I thought it would be helpful to share some of my favorite tips. Here are the top ten Construct tricks you’ll wish you knew sooner!
User Interface Hacks
1. Drag & Drop Extracts
Connecting to files takes a few steps because Construct remembers all of your previous data connections. Standard practice is to create a new connection or to select from existing connections. But there is a quicker way, especially when working with extracts like Excel and CSV files. Simply drag and drop connections from a folder directly into the connections window.
Depending on the situation, this can be faster than browsing your files or existing connections in Construct. Drag-and-drop also reuses existing connections when available (instead of creating a duplicate connection to a folder you previously connected to), making for a more orderly connections list.
2. Use Temp Outs While Creating or Troubleshooting a Job
The Temp Out node can be added in any location throughout your workflow.
The Temp Out node’s advantage over the standard data output node is that you can run it without any required configuration. Temp Out makes it easy to test tasks you establish in your workflow as you’re building it. Use this “test as you go” approach to make sure that a newly created variable in a transform is turning out the way that you expect, or that a transpose restructured the data in the way you hoped.
Another advantage of the Temp Out node is how easily you can add or delete them throughout your workflow. Whenever a workflow’s output produces unexpected results, you can pinpoint the stage in the workflow where the data isn’t being handled properly by attaching a Temp Out node, running it, and then moving it down the workflow to the next area where the issue might lie.
3. Split/Branch Your Workflow
Users often design jobs with a single output since, in many cases, a job only needs to generate one output. However, jobs often require testing while you create them, and outputs sometimes have more than one intended audience. Branch your workflow to save time and avoid repeat work.
As you can see above, rather than create one job to satisfy a data request and a separate job to report at a higher-level, branch your workflow to create two reports in the same job.
Branching also helps when you want to test changes to your workflow or check something partway through.
The above job’s shaded area shows how you could quickly check for duplicates without disconnecting any nodes. In short, use branching to add a node wherever you need it while leaving the rest of your job intact.
4. Rename Nodes
You likely know about this Construct trick, but it bears repeating. It is good practice to rename any nodes in your job to make it easier to identify what each part of the process does. You may not know that you can use the Edit > Find option to get a list of nodes in your job, then double-click on an entry to locate a specific node.
5. Close All Data Views
If your view is cluttered with too many open data view tabs, go to Window > Close all Data Views in the top toolbar to instantaneously clean up your workspace.
Force Multipliers/Doing More with Less
6. Multi-Column Cleanse
You’ve likely used the Cleanse node at some point in your work. It is a handy node that allows you to replace or revise values within a column. At times, you may want to apply one of the Cleanse node’s functions to multiple fields. Most users select one field at a time and perform a single function, but there’s a faster way.
Let’s say you needed to use the trim function to remove any trailing or leading spaces within your columns. Instead of selecting each column individually, hold ctrl to select multiple columns at once, then apply the same function to all selected columns.
7. Reorder Multiple Columns
If you’ve ever moved a column 200 places up by clicking 200+ times to do so, you’ll appreciate this Construct trick.
Instead of selecting the column you want to move, highlight all the other columns in the list that should be moved down by left-clicking on the top column, holding shift, and left-clicking on the bottom-most column. Then move them all down together. You just saved yourself 200 clicks.
8. Select Multiple Aggregations
When you use the Aggregate node, you likely want to take multiple similar measures of several columns. There are icons on the toolbar that allow you to take the Min of all columns, for instance. But perhaps you want a set of values for a subset of columns. Instead of checking the box for each measure, highlight multiple fields and select the measure you want to apply across the subset.
*Note that this behavior is slightly different in the latest version of Construct. In previous versions, you could click and drag in the measures column and then check the box in the active window.
Transform Your Transforms
9. Create Local Variables
Here’s a more advanced trick. The Transform node allows you to create local variables, which are calculations that can be performed in a transform function and then reused later in the same function. For instance, let’s say you need to calculate the average of a series of columns. This function would work:
If you prefer to break this out into separate pieces and simplify the formula, you could first calculate the total and store that as a local variable to be reused. The syntax for creating a local variable is: [VarName]:=expression; This is what the function above would look like with a local variable:
This function works just like the first but performs the calculations in two separate steps. It may be that you need to reuse the [SUM] value multiple times throughout the formula. A local variable saves you from retyping ‘[A]+[B]+[C]’ each time.
10. Add Comments To A Function
A relatively new feature in the Transform node is the ability to add comments to your function. If you use complex functions, it might be a good idea to include comments to explain the function’s goal. Check out the Comment function (available in the Transform node’s function list) to see the available syntax.
Here is what a single line comment looks like:
Put Our Service to the Test
Don’t Hesitate To Reach Out to Support
You’ll run into speedbumps with any software, and in most cases, that means turning to message boards or google.
Rapid Insight handles support differently. Our analysts are always happy to answer any questions you have. Don’t hesitate to reach out to email@example.com. And stay tuned for more Deep Dive webinars!
Have any Construct tricks of your own? Please leave them in the comments below!