Hello, 

In versions 7.17.3 and 7.17.4, to resolve the error, it is necessary to disable the feature Feature-UseTypeCastExpressionValidationInProcess by adding the following line to the application's \Terrasoft.WebApp\web.config file:

xml

<add key="Feature-UseTypeCastExpressionValidationInProcess" value="false" />

As an alternative solution, the issue can also be resolved by explicitly specifying data type casting in the formula.

Conversions that may lead to precision loss—for example, the error message “An error occurred in the formula: Cannot convert type 'Double' to 'Int32'”—must be handled explicitly through proper type casting.

Thank you !

Hello!

Yes, what you're describing is fully achievable in Creatio using a simple business process.

Here’s how it works:

We can create a form inside Creatio (as a section or a page) with the necessary fields: dropdowns, radio buttons, and free-text inputs.

When your sales reps click "Submit", a business process is triggered.

This process will:

1. Read the entered data from the form.
    
2. Format the data into a structured email body (for example, using a table or plain text with labels).
    
3. Send the email to the defined recipient, using Email element. The email body can include HTML for styling if needed.

As an alternative, you can also create a printable form (a printable report template). This report can then be reused within a business process and sent as an attachment in the email - for example, as a structured summary of the submitted data.

You'd basically change the Parent of the object from BaseLookup to BaseEntity. However, doing that will have the following impact: BaseLookup gives the object Name and Description, BaseEntity does not have those. This could break a lot of things with the page and also with data in records already saved.

I've not tested making this change, so definitely try this out in a test or dev system before attempting any of this in prod. What I would probably do, before you make the change from BaseLookup to BaseEntity, first create new custom Name and Description columns, for example UsrName, etc (assuming these are being used). Then migrate all data from the lookup columns Name & Description to the new custom UsrName & UsrDescription. Then change the page to map to these new custom fields. Then, change the parent of the object to BaseEntity.

Maybe there is some way to override the check for CanManageLookups for your one object as an exception, but I've never looked into that. I assume you'd need to go the route outlined above.

Ryan

Hmm. Not sure that is possible. Assuming that the params aren't loaded until the process is started, but it won't start without the param being loaded. Some of the OOTB processes for cases check that after. For an example, you can see the process "Run process: Reopen case and notify assignee on receiving an answer regarding the case" that sort of works like this. It starts for any email connected to a case. Then the script task checks to see if it is from one of the incident mailboxes. However, that seems like a bad approach, adding the overhead of a process triggering for every single email (the difference of the OOTB one above is that it's only for emails already connected to cases). Might be a better approach to use an EntityEventListener. If you wanted to keep the ease of using a process to do the main work, you could have the listener code flag the email, then that flag is what really triggers the process.

Does not appear to be possible without code. The send email process element doesn't expose any properties for the header. 

If you used the code approach, you could make a generic process you'd use as a subprocess. Any process that needed to send email could use that one process. However, you'd lose the ability for an easy designer for composing the email in the process and would have to send HTML to the subprocess to use as the body (not an issue with the subprocess route specifically, but an issue with using code to send the email). 

If it worked for you, you could create the emails themselves as message templates. The subprocess would need to process that template as well to generate the body. At leas that way you could use an easy designer to create the messages instead of just forming the HTML.

Ryan

That works for me, but it does require not running the subprocesses in the background (which I can see from your screenshot that you're not)

Hello,

Thanks for your question!

By default, campaigns in Creatio do not have a built-in restriction on running over weekends — they execute based on the Default campaign execution frequency setting (e.g., every X minutes or hours). However, if you're noticing that emails aren’t being sent during weekends, a few things could be affecting this:

-Trigger emails used in campaigns may have a configured delivery window that excludes weekends. It's worth checking their delivery settings.

-If your environment is hosted in the cloud, there's a chance the issue is related to site inactivity. When there is no user activity (which can be typical over weekends), the site may go into sleep mode to conserve resources — and that can prevent campaigns from executing.

If you're still experiencing this behavior, we recommend submitting a support case with remote access enabled.

Hello, 

To convert a file read via the "Process file" system action into a Base64 string within a business process, you will use a Script Task. This allows you to perform the conversion using C# and then pass the resulting string to an external API.

var files = context.Process.FindFlowElementByName("ObjectFileProcessingUserTask1")  
  .GetPropertyValue("ObjectFiles") as ICompositeObjectList<ICompositeObject>;

if (files != null)   
{  
  foreach(var file in files)   
  {  
      if(file.TryGetValue<EntityFileLocator>("File", out EntityFileLocator fileLocator))   
      {  
          IFile fileItem = UserConnection.GetFile(fileLocator); // This now works due to the 'using' directives [11]  
          using (System.IO.Stream stream = fileItem.Read())   
          {  
              using (var memoryStream = new System.IO.MemoryStream())  
              {  
                  stream.CopyTo(memoryStream);  
                  byte[] content = memoryStream.ToArray();  
                  var encodedContent = Convert.ToBase64String(content);  
                  Set<string>("EncodedFile", encodedContent);  
              }  
          }  
      }  
  }  
}

return true;

In the Methods tab, add the following to the Usings list:

• Terrasoft.File

• Terrasoft.File.Abstractions
   

   

Hello,

Our R&D team has already started a task to add new phone masks functionality which will be implemented in 8.3.0 version release.

For now, as a workaround, we suggest disabling the mask check in the phone number (in the designer).


We apologize for any inconvenience this may cause. 
Thank you for choosing Creatio!

Thank you for the confirmation.

Hello,

Please refer to the following, below is an example of how many tokens each intent can use: 

More information about tokenization can be found here: 
1. https://learn.microsoft.com/en-us/dotnet/ai/conceptual/understanding-tokens
2. Actual visual representation of token counts from Open AI: https://platform.openai.com/tokenizer

Regarding token usage, specifically why does 1 question use so many tokens:

Creatio AI performs additional actions beyond the messages generated by the user in the chat. Let me explain with an example:   

If you ask Creatio AI to create a draft of an email, Creatio AI performs several steps: 

1. Search for available skills.
2. Read the skill prompt and available actions.
3. Read data from the page.
4. Check system settings to confirm that the message can be sent.
5. Generate a draft message and display it in the chat (the only step visible to the user).
6. Perform additional actions, such as opening a card, etc.  
    

Steps 1–4 are not visible in the user interface, but they also consume tokens, which explains why the token usage may appear higher than expected. 

Learn more: https://www.creatio.com/products/pricing

Best regards,
Ivan

Hi Ivan

Thanks for your reply. I have looked at the documentation and aware of this information. I was seeing if other users have experiences of your AI implementation. Real world usage is often different, and as our testing showed such different token usage, I was interested to capture this feedback.

Hi Mark, 

Same as you:  found massive usage tokens consumption for little use. 

Unless OpenAI and Azure AI Integration price lowers, for now, use of AI is on hold, until we can plug to other LLMs  [which is planned in later subversions of Creatio 8.3 I believe ], some of them being known to be much less expensive then open ai 4o .

Hi Damien,

Thanks for taking the time to reply. This is what I was worried about, as the promised ability to select your own LLM means we are locked into what I hope is the highest tier.

I do not want to tell our team that the AI function is going to be provided and by day 4 of the month, all credits are used up and we would be looking at a very large cost. However, if the benefits are genuine then it is a case of matching to the cost value achieved.

My worry was that the cost of the tokens being so high, makes the value hard to justify.

Same experience for me as what Damien shared. It's a great marketing tool, but real world use doesn't seem to add much benefit IMO, especially compared to the high token usage for even what seems to be simple requests.