Many organizations, especially banks, insurance companies and other organizations that value trust highly are very careful about how links are formatted in emails that are being sent out. The key is that links need to be “safe” which in essence means they need to have the correct domain. For example company Contoso, this means that the domain needs to end with “contoso.com” and could for instance be “forms.contoso.com” or something similar. However, there is currently no way to create “branded” links in CI-J and this feature was deprioritized and removed from Release wave 2 2024. I (and a some customers of mine) are hoping it will be coming soon. In the meantime, I have been considering what options there are. Here is one, with a simple webpage that just redirects to some other webpage based on some parameters.
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Redirecting...</title>
<!--
This webpage will redirect the browser to the url indicated in the query string paramater "url", forward all other query string parameters
and also track the webusage in Dynamics 365 CIJ
Ex. url (file located in c:\temp): file:///C:/temp/redirect.html?url=https%3A//skandia.se¶m1=value1¶m2=value2
-->
<script>
// Function to get query string parameters
function getQueryParam(name) {
const urlParams = new URLSearchParams(window.location.search);
return urlParams.get(name);
}
// Function to get all query string parameters except 'url'
function getOtherQueryParams() {
const urlParams = new URLSearchParams(window.location.search);
let params = [];
urlParams.forEach((value, key) => {
if (key !== 'url') {
params.push(`${key}=${encodeURIComponent(value)}`);
}
});
return params.join('&');
}
// Redirect to the URL specified in the 'url' query parameter
window.onload = function () {
const targetUrl = getQueryParam('url');
if (targetUrl) {
// Basic validation to ensure it's a valid URL
try {
const validatedUrl = new URL(targetUrl);
// Get other query parameters and append them to the target URL
const otherParams = getOtherQueryParams();
if (otherParams) {
validatedUrl.search = validatedUrl.search ? validatedUrl.search + '&' + otherParams : '?' + otherParams;
}
// Perform the redirect
window.location.href = validatedUrl.href;
} catch (e) {
console.error('Invalid URL:', targetUrl);
document.body.innerHTML = '<h1>Invalid URL</h1><p>Please provide a valid URL as a query parameter.</p>';
}
} else {
document.body.innerHTML = '<h1>No URL Provided</h1><p>Add a "url" query parameter to redirect.</p>';
}
}
</script>
</head>
<body>
<p>Redirecting...</p>
</body>
</html>
The code simply parses the parameter url and redirects to this url including all other query string parameters. This is important as that will maintain any UTM tags that were added.
The point is hence to place this webpage on any page internally at the company, for ex. https://contoso.com/redirect.html and then call it with the parameters indicated like https://contoso.com/redirect.html?url=https%3A//crmkonsulterna.se¶m1=value1¶m2=value2
In CI-J this means that you either manually have to assemble these url:s or create some function, preferably connected to the ribbon to easier be able to add this redirect url to the email.
Tracking
As we need to make sure that the URL is not using the built in tracking in CI-J, using the normal button functionality of “Tracking” needs to be switched off, as the tracking functionality uses Microsoft redirect pages and hence break the typical “no link outside our domains”-policy.
We do need to track anyway and there are two ways that I can see that off the top of my head will achieve this:
Using webtracking
Using custom triggers
Webtracking is, at the time of writing this article, a preview functionality, and we might hence be a bit hesitant in trying it out. If you want to, there is a setting switch for it under settings in CI-J. Then reload the page and click on “Web tracking” to get the correct script.
Custom triggers has been around a while and is probably easier to use. Create a custom trigger and get the code snippet for it and add it to the redirect-page. Every time someone hits the redirect page the custom trigger will be called. When using triggers, I would recommend using parameters to forward all relevant data to the journey where you want it.
I hope this helps you or gives you inspiration on how to do something similar!
In a previous post I blogged about how to break down the Form script that was exported from CI-J (Real Time). As some customers asked me about making this into a script that was the same and was dynamic based on query string parameters (parameters in the URL), I worked a bit on that and thought I’d share it here;
A few things that can be mentioned. This script expects that there will be two query string parameters: id – the id of the form. Click on the form and copy it from the URL after the “id=” orgid = the orgid which you can find in the PPAC or in the export of the script as I described in the previous article. If it is placed on the url: https://contoso.com/form.html then an example of the url would be:
Finally some organizations also have problems with loading a script from external sources. I will look at that too. Mainly there are two option. Copy-paste the entire script inline into this script or copy the script file to an “internal” or recognized store with a public URL and change the src-attribute. If you move this away from referencing Microsoft, I would recommend checking their website on a regular basis to make sure it hasn’t changed.
I have recently been working with a customer with a large (500+ GB dataverse db) production instance and we are attempting to cut this down quite a lot. The natural way of doing this is bulk delete but it seems that if the underlying SQL isn’t up to speed, your jobs might end without actually being done.
My customer has a production instance of over 500GB which is costing them some substantial money and hence we started looking at removing some of the less necessary data. In this case it was the order products (salesorderdetail) which we have perceived that we do not need more than one year after delivery. Hence we set up an advanced find, agreed on the exact filters, saved the view and then tried to remove the records (well over 5M) using bulk delete. I restarted it several times. You can see the chronological results in the screenshot below:
As you can see, sometimes it actually deleted a few records, but most times, it didn’t. My professional analysis of this (=guess) is that this is caused by bulk delete not handling exceptions like SQL Timeout properly or that it has a limit on the number of times it will retry.
I also, in parallell tried to remove the same records with the same FetchXML using SSIS/Kingswaysoft and here I have several times gotten this error. I have had to turn down the knobs to a very low setting to get it to work, but the error message I did get was:
(Status Reason: ServiceUnavailable): The remote server returned an error: (503) Server Unavailable.”.
As you can see, there is a lot of noise but it clearly, in the highlighted part, say that there is a SQL Timeout.
Hence my takeaway from this is that you need to be a bit wary of bulk delete in large instances or in general as it might indicate that it has completed successfully but in fact it stopped due to SQL Timeout (or some other platform related issue).
On a personal note, I really hope the bulk delete functionality gets a modernized revamp soon. It is really old and is becoming more and more relevant.
Recently I needed to get the marketing insight replicated data from the data lake where it is replicated to, to Customer Insight. However, that turned out to not be very easy as the data had some formatting issues. However, I found that the Marketing reports that Microsoft have release for Power BI can be used as inspiration for how to query the data.
I am currently working on a rather long Proof of concept of Customer Insight for a customers. One of the things they wanted to see if it worked to get into Customer Insight was EmailClicked which is part of the Marketing Insight data that Marketing stores in an internal database. This can rather easily be replicated to an external datalake, with some configuration in the admin tab in Marketing. However, when I tried to connect to that data, using the Azure Datalake gen2 adapter in Customer Insight, it consistently said that there was no data. After some digging I found that the main reason for this was that the files in the datalake did not have the propper file-ending for the datalake connector to understand them. In short, they are csv-files, but do not have the filename xyz.csv. Simple enought problem, I thought but as I am not super comfortable working with datalake data, I tried to figure out some way of easier solving this issue. First I tried using the dataflow connector to ADLS gen 2 but that got the same problem. Just so you know.
Then it struck me, data flows use PowerQuery/M which is the same thing that is used in Power BI, AND Microsoft have release some marketing reports for Power BI that utilize the data in the datalake combined with dataverse. I hence opened one of these and tried to copy the entire data query part. It turned out to be more than 20 different components. Datasources, configuration, functions and more. When pasted into the Data Flow, it simply didn’t work and as usual, I didn’t get a very good errormessage. But my hunch was that maybe the logic is too complicated with too many internal connections for the data flow. If you know the limitations here, please leave a comment. This didn’t stop me, my next step was to remove all unnecessary stuff from it, like the dataverse queries, config and such, but still, it didn’t work. So I attempted to move it all into one single M-script with a hardcoded referece to EmailClicked. When I did that, it worked! This is the final result:
let
Source = AzureStorage.Blobs("<datalakename>"),
#"ContainerContent" = Source{[Name="<containername>"]}[Data],
#"Removed Other Columns" = Table.SelectColumns(ContainerContent,{"Content", "Name", "Date modified", "Attributes"}),
#"Filtered Rows" = Table.SelectRows(#"Removed Other Columns", each [Name] <> "model.json"),
#"Sorted Rows" = Table.Sort(#"Filtered Rows",{{"Date modified", Order.Descending}}),
#"Expanded Attributes" = Table.ExpandRecordColumn(#"Sorted Rows", "Attributes", {"Size"}, {"Size"}),
#"File Name column" = Table.DuplicateColumn(#"Expanded Attributes", "Name", "File Name"),
#"Remove csv" = Table.ReplaceValue(#"File Name column","csv/","",Replacer.ReplaceText,{"File Name"}),
#"Split Column by Delimiter" = Table.SplitColumn(#"Remove csv", "File Name", Splitter.SplitTextByEachDelimiter({"/"}, QuoteStyle.Csv, true), {"Interaction Name", "File Name"}),
#"Transform" = Table.TransformColumnTypes(#"Split Column by Delimiter",{{"Interaction Name", type text}, {"File Name", type text}, {"Size", Int64.Type}}),
#"Add Datestamp" = Table.DuplicateColumn(#"Transform", "Date modified", "Datestamp"),
#"DateStampFormat" = Table.TransformColumnTypes(#"Add Datestamp",{{"Datestamp", type date}}),
TodayFunction = DateTime.FixedLocalNow,
#"Add Today" = Table.AddColumn(#"DateStampFormat", "Today", each TodayFunction()),
#"Changed TodayType" = Table.TransformColumnTypes(#"Add Today",{{"Today", type date}}),
#"Add DaysFromToday" = Table.AddColumn(#"Changed TodayType", "DaysFromToday", each [Datestamp]-[Today]),
#"Changed DaysFromToday" = Table.TransformColumnTypes(#"Add DaysFromToday",{{"DaysFromToday", Int64.Type}}),
Result = Table.RemoveColumns(#"Changed DaysFromToday", "Today"),
result2 = Table.SelectRows(Result, each [DaysFromToday] >= -180),
FilteredByInteraction = Table.SelectRows(result2, each [Interaction Name] = "EmailClicked"),
#"AddFileContents" = Table.AddColumn(#"FilteredByInteraction", "FileContent", each
Table.PromoteHeaders(Csv.Document([Content],[Delimiter=",", Encoding=1252, QuoteStyle=QuoteStyle.Csv]), [PromoteAllScalars=true])),
#"ContentTable" = Table.SelectColumns(AddFileContents,{"FileContent"}),
#"NoDataFiles" = Table.IsEmpty(#"ContentTable"),
InteractionTable = Table.ExpandTableColumn(#"ContentTable", "FileContent", Table.ColumnNames(ContentTable{0}[FileContent])),
#"Transformed" = Table.TransformColumnTypes(InteractionTable,{{"Timestamp", type datetimezone}}, "en-US"),
#"Duplicated Column" = Table.DuplicateColumn(#"Transformed", "Timestamp", "Datestamp"),
#"Datestamp" = Table.TransformColumns(#"Duplicated Column",{{"Datestamp", DateTime.Date, type date}}),
#"RenameId" = Table.RenameColumns(#"Datestamp",{{"InternalMarketingInteractionId", "Id"}}),
#"keyedtable" = Table.AddKey(#"RenameId", {"Id"}, true)
in
#"keyedtable"
I have highlighted the datalake name, container and the table. I hope this helps if you are having a similar issue!
Have you ever tried creating bulk deletes in dataverse/Dynamics 365? It is still a very old interface and it is hard to control the exact definition of what is to be deleted as you cannot see the actual FetchXML that is being used. It is also hard to see existing recurring bulk delete jobs and what FetchXML they are using. Based on these facts, my colleague Ebba Linnea Nilsson and I decided to make our first plugin for XrmToolBox (XTB). We are now proud to announce that it is released and available for free (as usual in XTB). We have also identified a bug in the platform related to this, which I will describe below.
The bulk delete functionality in dataverse and hence in Dynamics 365 CE is an essential function for many organizations. This is especially true since GDPR was introduced and there is a strong legislative requirement to remove personal data that cannot be justified to be stored. There are also other reasons why the bulk delete functionality is more and more important and that is based on the capacity costs that can be inferred on a Power Platform tennant. Firstly just storing data, especially in the database storage in dataverse has a non trivial cost at $40/GB. There is a lot of value per GB, so it might not be justified to say that it is too expensive but removing unnecessary data is definitely something that can be worthwhile especially for larger implementations. I personally work with a customer in the travel retail industry which has millions of customers and some tables have 40M+ records. There are also a lot of integrations and automations causing a lot of data to be created. Data that at some point needs to be removed. As all data should, or there should at least be a conscious decision not to remove it and why if so. As you might be aware, if you have read my other articles in this blog, I have previously used SSIS and Kingswaysoft to remove massive amounts of data. However, now that the API Entitlements will be introduced (6 months after the report for API Entitlements is made Generally Available), we need to start to become more and more restrictive in using the APIs for massive data management, like deletes. Hence, we try to move as much as possible to the built in Bulk Delete.
Working with the built in bulk delete functionality is a bit sad. It is very old and you have to click through a wizard kind of experience to be able to set them up. But the most limiting factor is that you cannot see the actual FetchXML of an active recurring bulk delete and you cannot input FetchXML directly into the bulk delete.
The Options page of the Bulk Delete Wizard
Having worked with this wizard you might also have noticed something a bit off. If you create a view which shows all contacts that have no activities. Then this will work when using it in the system like a view. However, if you try to use this view in a bulk delete it will “simplify” this and remove that part of the query. My thought that this was a limitation in the UI based on the fact that it is very old and hasn’t gotten any love for many years (decades). My assumption was hence that creating a bulk delete via the API would allow me to create bulk deletes that were based on FetchXMLs that you couldn’t even input from the UI. These were the reasons for us starting to create this plugin and it was so useful that I used it in debug mode for several weeks before finalizing it and publishing it.
Now we have released the first version and you can download it directly from XTB. I would like to give a huge thanks to Jonas Rapp who helped us out a lot with both connection to his tool FetchXML Builder but also the general setup of the plugins and the details of getting it approved as a Plugin for XTB.
If you have any suggestions, comments or otherwise, leave them on the GitHub repo https://github.com/crmgustaf/BulkDeleteManager or down below. We already have a bunch of stuff we want to do. Ah, yes, and the bug we found, it seems that the outer joins that was a rather recent addon to FetchXML is not supported by the actual platform. Hence the UI and the platform match in that perspective. Just to make it clear, what happens is that you input a FetchXML saying “All contacts with no activities” or something like that, which it will simplify to “All contacts” which is not really what you want.
As it is supported to create bulk delete jobs via the API, I do think that this still can be seen as a bug as there is no clear documentation on this or even a control when creating the bulk delete job with a FetchXML that will be incorrectly parsed. My suggestion is hence to implement the new FetchXML parser in Bulk Delete functionality, at least on the platform side. With the current setup, it is very possible that bulk deletes are created that remove a lot more than what was initially intended which can be very damaging to any organization. And from a GDPR perspective, this type of query is rather common, as it might be definied that contacts with no cases can only be stored for 2 years, but with cases for 10. To remove the ones without the cases, you would make the question “Remove all contacts with no cases with created on > 2 years”. This would then cause all contacts with created on > 2 years to be removed regardless of if it has a case or not.
To inform users of this, we have added a warning, every time a new bulk delete job is to be saved. Hopefully Microsoft will fix this soon.
Recent Comments