Duration and frequency


Setting duration and frequency are critical to quantifying models in LINQ, as it allows you to specify how long actions take and how frequently they occur.  By setting these values for each action node, you can: 

  • Determine the time required for activities and their frequency,
  • Gain insights into workflow efficiency,
  • Enable data flow through your model,
  • Track resources accurately,
  • Optimize processes effectively.


Duration refers to how long each action takes. Once frequency is set, you define how much time each action consumes. For instance, if the validation of an account takes 5 minutes per record, you set the duration for that specific action.



Frequency defines how many times an action occurs—such as the number of customer accounts being created annually. For example, if an action node represents account creation, and we know that this occurs 3,650 times a year, we set the frequency accordingly. 


It is advisable to use the smallest period of duration you can get data for, i.e. a per day frequency is more advantagous than a per year frequency, even though LINQ annualises the data. This is because we can more easily update frequencies as things change as we don't have to calculate annual volumes - LINQ will do that for us.


To set duration and frequency for an action node in LINQ, you'll need to use the Properties Panel in the Canvas or double-click each action node. 


Properties Panel


The Properties Panel in LINQ offers detailed settings for the selected node, or nodes, allowing you to define crucial parameters like duration and frequency for your action nodes. 


 

Here are the features of the Properties Panel in detail: 


  • Select mini graph: It allows you to view either an overview or a subset of the model based on the selected node.  By choosing from options such as Adjacent, Context, Downstream, and Upstream, you can focus on specific areas of the model based on the selected node. 


Note: To view a subset of the model, the node must be selected within the model.


  •  Zoom to node: It allows you to zoom in on the selected node within the model. 


  • Node details: This enables you to edit the specifics of the selected node.  

The node details are divided into the following categories: 

  1. Name: Input or edit the node's name to describe the node.
  2. Description: Provide additional context or details about the node. 
  3. URL: Optionally include a URL link for further reference. 
  4. Type: Select the specific node type based on the selected node. The dropdown menu dynamically updates depending on the type of node you're editing. For instance, if an Action node is selected, only relevant Action node options will be displayed. 
  5. Notes: Add any supplementary information about the node. 
  6. Tags: If tag groups have been created, you can apply tags to the selected nodes. 


Note: All changes will remain unsaved until you save the model.


  • Connect node to other models: Checking this will allow the information node to be used as an input in your other models. For more details on how to connect models, please refer to the Advanced Modelling section. 


Quantifying the Model 


To start quantifying your model, follow this step-by-step process:


Step 1: Start with the Capture Node


The Capture Action is the starting point of your process, introducing the initial volume of work into the model and influencing downstream calculations. Setting the Capture Action correctly is fundamental for several reasons:

  • Establishing the workload foundation: The Capture Action determines how many items or "work units" flow through the model. 
  • Driving downstream calculations: By setting the Capture Action's frequency accurately, subsequent actions can use the input frequency to automatically determine how often they occur.  This eliminates the need to manually calculate the frequency for each action and ensures consistency throughout the model. 

  • Simulating scenarios and understanding impact: Once the Capture Action is set, you can use it to run what-if scenarios. For example, what would happen if the workload doubled? By simply increasing the frequency in the Capture Action, you’ll immediately see the impact on the entire process. 
  • Enabling precise cost and resource analysis: Because the Capture Action sets the pace of the workflow, LINQ can calculate how much time and resources are needed at every stage of the process. This helps you understand the overall cost of each action and where you might be able to optimize. 


The following example shows how to quantify a Capture node. You can work through the example by watching our YouTube videos "LINQv2 Onboarding Video - Quantifying your model" and "LINQv2 Onboarding Video - Quantifying Action Duration and Frequency-Advanced". 


Let's use Create new account as the Capture node in the model. To begin, you can either double-click the Capture node or select it and scroll down in the Properties Panel on the left-hand side, and then follow these steps:


1. Click Missing frequency, click here to add one to open the Frequency Dialogue Box.  

  • In this example, Create new account takes 10 minutes (how long does the action take) to complete. 
  • It occurs 10 times (how often is the action taken) each workday, and 52 weeks in a year.  You can specify the frequency in various units like per day, per workday, or per week, and adjust the number of weeks in a year to account for holidays or other non-working periods.

 

2. Click Save. The Frequency will now be set as "10 each work day with each taking 10 minutes" in the Properties Panel on the left-hand side.  

  • Setting the frequency to 10 times per workday for a year results in 2,600 information inputs each of Name, Address, and Email address in the model.
  • If you would like to customize which labels are visible on your model, click the Show labels icon located at the top of the toolbar. In this case, select Processed, which will now display the number of items processed through the model based on the set duration and frequency.


3. As Create new account action produce multiple outputs (like Name, Address, Email address), you can use Information outputs to control the percentage of each outputs allows you to model different process flows effectively.  This is especially useful for scenarios where some outputs might trigger rework loops or lead to further actions with varying frequencies.  In this case here, the output percentages for all associated information outputs are all 100%.  

 

Step 2: Set Duration and Frequency for the Subsequent Actions 


For actions following the Capture action, you can leverage the input frequency from the preceding action.  In this case here, after Create new account as the Capture Action, we now move on to the subsequent actions which are Validate the account, Fix the errors, Research and fix errors and Load into CRM.  Here's how:


1. When an action node has multiple information inputs (from different preceding actions include Capture Action), you can use the Information processing drop-down menu, select how the action will handle multiple information inputs.  This allows you to consider many information inputs as a single group, or act on each information input separately, without having to duplicate the action. Use the Information Input dropown to select the appropriate method. 

LINQ provides four flexible options: 

  • Take the sum: Adds up all the input numbers, useful for actions that need to be performed individual for each input. 
  • Take the greatest: Select the highest input number.
  • Take the least: Select the lowest input number, appropriate when different inputs are processed together as a single unit. 
  • Take the average: Calculates the average number of inputs.  


Note: This information processing can only be applied to actions after the Capture Node, and every time there are two or more information inputs into an action node, unless you manually over-ride the action frequency, in which case this option is not available.  


In this case, Create new account results processing information for Name, Address, and Email address. Each piece of information is now an information input for the next subsequent action of Validate the account.

 

Let's say Validate the account requires individual processing as a single entity from Name, Address and Email address information inputs, so we select Take the least information inputs is more suitable. By selecting Take the least information inputs results Validate the account as 2,600 information inputs as the least in the model.


2. For each subsequent action, click Missing frequency, click here to add one, you'll notice a toggle bar indicating you're using the calculated input frequency from the previous action (which is 2600 per annum from the Information processing defined as Take the least information input). This will allow you automatically calculate frequencies based on the Capture action.  In this example, let's say Validate the account will take 5 minutes by using input frequency of 2600 per annum.    

 

However, if you want to override the calculated value, manual override of input frequency is an option. This is useful for scenarios where an action's frequency is independent of the calculated flow. 


Note: If you manually override the frequency in any subsequent action, the Information Process options/drop down will no longer be available because the frequency is now set manually.   



3. Click Save. Then the statement of Frequency will change to green color.  


4. Set Information output percentages for two information outputs: Valid account and Invalid account. To reflect a more realistic scenario, you can adjust the information output percentage from 100%:

  • Valid account: 80% - Indicating that 80% of the accounts are validated successfully. 
  • Invalid account: 20%- Indicating that 20% of the accounts require further action. 


5. After adjusting the percentage, the number of items processed through each output will automatically update based on the input frequency.  In this case, if the validation action processes 2,600 accounts, 80% (2,080) will be valid, and 20% (520) will be invalid.  


6. To set up frequency for the Fix the error, Research to fix error and Load into CRM actions, let's repeating the same steps:

  • Set the Information processing: Take average of information inputs for Fix the error and Research and fix errors to indicate it takes an average time to fix errors and do the research. And Take sum of information inputs for Load into CRM action, which is the sum of information inputs from Validate account and Valid account information inputs/nodes. 


Note: You can't set the frequency and information processing for multiple actions at the same time. You will need to hover over each action node individually and set the frequency and information processing for each action separately.  


  • Click Missing frequency, click here to add one for each action, for Fix the error, let's use an input frequency of 520 per annum, and takes 10 minutes for this action. Research and fix error will take input frequency of 104 per annum and the action duration is 5 minutes.  Load into CRM will take input frequency of 2600 per annum and set the action during to 1 minutes. 


Note: Remember to Save the changes, each of action must be done individually.  


  • Set up the information outputs for each action as follows:
    1. For Fix the errors: 80% validated accounts (416 processed), 20% invalidated accounts (104 processed)
    2. For Research and fix errors and Load into CRM: Both have a single information output (Validated account and Customer Account).  Keep these at 100% will result 104 processed for Validated account, 2600 processed for Customer account.


To summarise, building an accurate LINQ model requires:

  • Accurate duration and frequency settings: Carefully define the duration and frequency of each action, especially for the Capture action, to accurately represent your model process. 
  • Appropriate information processing options: Choose the right Information Processing option based on how the action handles multiple information inputs. 
  • Controlled Information Outputs: Set the percentages for each Information Output to control workflow through different pathways. 


By focusing on these elements, you can create a powerful and insightful LINQ model that accurately reflects your processes.  To view how to quantify action duration and frequency, please refer to YouTube videos "LINQv2 Onboarding Video - Quantifying your model" and "LINQv2 Onboarding Video - Quantifying Action Duration and Frequency-Advanced".