In this post, we’ll take a closer look at how ChatGPT can assist you in finding the answers you need while navigating the Topliners community. While ChatGPT is an excellent tool for responding to straightforward questions, the jury is still out on its ability to address more intricate queries that even experienced users may struggle with.
Oracle Eloqua Topliners community connects thousands of individuals who implement Oracle Eloqua in marketing campaigns automation. This online forum isa valuable resource for finding answers and insights from fellow users and industry experts. But what happens when you can’t find the answer you’re looking for on Topliners? That’s where the omniscient ChatGPT might be of help. This AI-powered language model is trained to understand natural language queries and provide relevant, accurate responses in seconds.
Prompt: Are clicks on the unsubscribe link in an email counted in the click-through totals and rates in Eloqua?
ChatGPT: Typically, clicks on the unsubscribe link in an email are not counted in the click-through totals and rates in Eloqua or any other email marketing platform. The purpose of the unsubscribe link is to allow recipients to opt-out of receiving further emails from the sender, so including clicks on this link in the click-through totals and rates could be misleading and inflate the metrics. Additionally, including clicks on the unsubscribe link in the click-through rates could negatively impact the sender’s reputation and deliverability, as it may indicate to email providers that the sender is not respecting recipients’ preferences to stop receiving emails. Therefore, it is best practice to exclude clicks on the unsubscribe link from the click-through totals and rates in Eloqua or any other email marketing platform.
Our AI interlocutor was able to provide a clear and concise answer, explaining that typically clicks on the unsubscribe link are not counted in the click-through totals and rates in Eloqua or any other email marketing platform.
Here’s a link to a Oracle Base Knowledge Base article that confirms the above: https://support.oracle.com/epmos/faces/DocumentDisplay?id=2448112.1
Our comment: ChatGPT has demonstrated its ability to provide accurate and detailed answers to basic requests, making it a valuable tool for users seeking assistance with Oracle products and services.
Our score: 10 out of 10 (excellent!)
Topliners Query: Can user accounts with asset dependencies be deleted?
Prompt: How do I add or remove users from my Eloqua account?
ChatGPT: To add or remove users from your Eloqua account, you need to have administrative access to your account. If you don’t have administrative access, you’ll need to contact someone who does to make these changes.
To add a user to your Eloqua account:
To remove a user from your Eloqua account:
It’s important to note that when you remove a user from your Eloqua account, all of their associated assets (emails, landing pages, campaigns, etc.) will be deleted as well, unless they have been shared with other users or groups. Therefore, it’s a good idea to review the user’s assets before deleting their account to make sure that any important assets are not lost.
Our comment: the answer is correct, although… the last paragraph is not entirely accurate. When you remove a user from an Eloqua account, their assets are not automatically deleted. However, the assets will no longer be accessible to other users unless they have been shared with them.
Our score: 8 out of 10 (pretty good!)
Topliners query was about possible ways to identify the web page from which an Eloqua form submission is happening.
Prompt: I have an Eloqua form which is being used across multiple sites. What are the possible ways to identify the web page from which the Eloqua form submission ishappening?
ChatGPT: There are a few ways to identify the web page from which the Eloqua form submission is happening:
Prompt: Can you please provide me with an example JavaScript that can be used to populate the hidden field with the URL when the form is loaded on the page?
ChatGPT: Sure, here is an example JavaScript code that you can use to populate a hidden field with the URL of the web page from which the Eloqua form was loaded:
In this example, you would need to add a hidden input field to your Eloqua form with an ID of “hidden-url-field” (or whatever ID you prefer), like so:
When the Eloqua form is loaded on a web page, the JavaScript code will retrieve the URL of the web page using the window.location.href property, find the hidden in put field by its ID, and set the value of the hidden field to the URL. When the form is submitted, the value of the hidden field will be included in the form data, allowing you to identify the web page from which the form was submitted.
We were particularly interested in implementing solution #2, which involved using JavaScript to populate a hidden form field with the URL of the web page from which the form was loaded. ChatGPT provided an example JavaScript code which could be used to accomplish this, and we were happy to confirm that the solution worked successfully.
Good job, ChatGPT!
Our comment: we were very impressed with how ChatGPT handled the query and provided a helpful solution. This is a great example of how AI can be used to provide assistance and solve problems for users in a variety of contexts(particularly technical ones).
Our score: 10 out of 10 (excellent!)
Prompt: I need to export the lead score for all the contacts that we have in our Eloqua database for further analysis.
ChatGPT: To export lead scores for all contacts in your Eloqua database, you can follow these steps:
Prompt: Regrettably, the Lead Score fields cannot be found in the Contacts Table, making it impossible to incorporate them into the Segment…
In this case, we asked how to export lead scores for all contacts in their Eloqua database for further analysis. The initial response provided by ChatGPT suggested exporting the lead score field, but we corrected our AI partner saying that the lead score field was not available in the Contacts Table.
To solve the issue, ChatGPT suggested exporting related fields and calculating a lead score for each contact using a spreadsheet program. While this solution may not be as accurate as exporting the actual lead score field, it can still provide valuable insights into the engagement levels of contacts.
However, we then asked ChatGPT if it’s possible to use Eloqua Insight to get more in-depth information about lead scoring. ChatGPT correctly suggested using Eloqua Insight to create custom reports and dashboards based on Eloqua data.
More information about how to check the performance of your lead scoring model can be found here: https://docs.oracle.com/en/cloud/saas/marketing/eloqua-user/Help/LeadScoring/Tasks/ViewingThePerformanceOfYourLeadScoreModel.htm
Our comment: this case exemplifies ChatGPT’s limitations in handling complex requests without sufficient detail or context. It also highlights the significance of complementing AI technology with human expertise to provide the most accurate and relevant answers.
Our score: 3 out of 10 (not particularly helpful)
In this case, a member of Topliners community asked about a possible workaround for collecting approval history for their clients' internal audits in Eloqua. The user had already investigated the Eloqua Approval tab, Audit Log, and different Analyses in Insight, but couldn’t find a solution.
Prompt: I’ve recently been working with clients who have a requirement on collecting approval history for their internal audit purpose. Currently, users can view the history of all campaign approvals on the homepage screen of Eloqua under the Approval tab but aren’t able to find a report to export this history anywhere. I’ve looked into the Audit Log as well as different Analyses in Insight but couldn’t find a solution. Does anyone have similar experiences and can share possible workarounds for this?
ChatGPT: One possible workaround for this could be to use Eloqua’s API to extract the approval history data and generate a report. The API can be used to fetch the approval history data for specific campaigns, as well as for specific time periods. You can use this data to generate a report in a format that meets your clients’ requirements.
To get started with using the Eloqua API, you can refer to the Eloqua REST API documentation. You’ll need to authenticate your requests using an API key, which you can generate from the Eloqua Administration page.
Once you have authenticated your requests, you can use the GET /assets/email/{id}/approvals endpoint to fetch the approval history for a specific email campaign. You can also use the GET /assets/campaign/{id}/approvals endpoint to fetch the approval history for a specific campaign.
Once you have fetched the approval history data using the API, you can use a scripting language such as Python or a data manipulation tool like Excel or Google Sheets to process the data and generate a report in the desired format.
Alternatively, you can consider using a third-party reporting tool that integrates with Eloqua and provides more advanced reporting features than the built-in Eloqua reports. Some examples of such tools include Looker, Domo, and Tableau.
ChatGPT provided some initial solutions to extract approval history data for campaigns in Eloqua, such as using the Eloqua API and the Bulk API. However, when we validated the response against the official documentation, the specific endpoint suggested by ChatGPT for fetching approval history data was not available, neither in REST nor Bulk API. When we suggested using the Eloqua Firehose Service (which is the correct answer to this question) to extract the data, our AI conversation partner did not provide any further guidance on this solution.
Our comment: in this case, expertise of a technical Eloqua consultant (human) was necessary to answer the question as the available Eloqua API documentation did not provide sufficient detail to enable ChatGPT to distinguish between two different status fields. While ChatGPT initially provided some recommendations (irrelevant), it was ultimately the input from a human that helped to identify a suitable solution to the problem.
Our score: 0 out of 10 (misleading/irrelevant)
While the AI assistant provided generally accurate information for common Eloqua questions, the Topliners community contains more specific implementation details and nuances from Eloqua experts and power users. AI can serve as a good starting point, but the forum remains an invaluable resource.
We hope you found this post on ChatGPT and Oracle Eloqua Topliners helpful and informative! At Ivenium Marketing we’re always looking for ways to support Marketing Automation teams, and we believe that ChatGPT (and other generative and transformative AI tools) can be a game-changer when it comes to finding quick, accurate answers to your questions.
We’d love to hear your thoughts on this post and the topics we covered. Did you find ChatGPT’s capabilities impressive? Have you had any experience using ChatGPT or other AI tools in marketing automation?
If you’re interested in exploring this topic more deeply, be sure to check out our upcoming blog posts on AI and its capabilities for the Marketing Automation industry. In our next posts, we’ll continue the interrogation of ChatGPT in various areas of marketing automation: Data Standardization processes, troubleshooting HTML code, fine-tuning Lead Scoring Models and other topics. We’ll cover Oracle Eloqua, Adobe Marketo Engage, Hubspot, Salesforce Marketing Cloud, Adobe Campaign, Salesforce Pardot.
Stay tuned for more content from us!