Setting Up Fallback to Re-Run Failed Column Associations

Loh Y. · 2025-06-04T03:09:22.700Z

Hi there, how can I setup fallback to re-run this column whenever it failed to find the association for the first time? I would like to run it again perhaps after 12 hours. For now, I've manually force run for the rows where the association was not found previously. Workbook .

Setting Up Fallback to Re-Run Failed Column Associations | Clay

10 comments

· Sorted by Oldest

Channeled
APP
·
·
You can use a combination of scheduled columns and conditional runs to create this fallback system: 1. Set up a scheduled column to run daily using the table settings (⛭ icon) > "Run the table" > "On a schedule" > select "Day". 2. Create a conditional run formula for your column that only processes rows where the association wasn't found. You can use an AI formula that checks if the result is empty or has an error. This approach will automatically retry failed associations every 24 hours. While we don't have an exact 12-hour option, the daily schedule is the closest available timing.
Sources: AI Formulas + Conditional Runs [Automated Outbound] - Clay University Datagma integration overview | Documentation | Clay University Scheduled columns | Documentation | Clay University
Loh Y.
·
·
I'm able to do this. But how can i set formula to only run on rows where association wasn't found?
Channeled
APP
·
·
To run your column only on rows where the association wasn't found: 1. Click on your column's "Run Settings" 2. Select "Conditional Run" 3. Click the "Use AI" button 4. Enter something like: "Only run if the [association column] is empty or null" The AI will generate a formula that checks if your association field is blank, allowing the column to only run on rows where the previous attempt failed.
Sources: AI Formulas + Conditional Runs [CRM Enrichment] - Clay University AI Formulas + Conditional Runs [Clay 101] - Clay University Guide: Ways to save Clay credits | Documentation | Clay University
Loh Y.
·
·
Please help. I need to re-run that column after the 1st run and only the association was not found in the first place.
Loh Y.
·
·
For this setting, I'm unsure if there is any implication to re-run rows where the association was already found in the first place and I would prefer for us to not run as I don't want to risk calling too much of the HubSpot api
Loh Y.
·
·
But if you think re-runnign the already successful rows would not cause any problem - do let me know and also your reasoning.
LuisArturo
·
·
Hey there Loh thanks for reaching out, sorry for the long delay here, to clarify currently there is no way to have the schedule column run only run for columns that did not find a results, although this is something that the team is looking into. In this case you could go ahead and rerun this HubSpot integration as there is no credits involved with rerunning this column, as the only thing it would do is just lookup whatever data there is in regard to the Associated company objects
Loh Y.
·
·
Are you saying it's okay to turn this setting on? My question next would be what if an association is already found? Would they try to find the association again?
Loh Y.
·
·
Cause at the point where the row is triggered to find the company association, most of them would just have 1 company associated. But there will be a point in the future where a contact can be associated to 2 or more companies. This don't always happen, but still I'm simply afraid that by re-running the retrieving company association step would cause the past rows to find new company object and then it would re-update the row again - which is not a behavior I want.
LuisArturo
·
·
Sorry for the delay again, if this integration was to run again it would try the find the association again. What you can do here is then is for any integrations down stream from this one you can go and turn off the Auto-update setting inside of the column setup menu
(11 kB)

Channeled
APP
·
·
You can use a combination of scheduled columns and conditional runs to create this fallback system: 1. Set up a scheduled column to run daily using the table settings (⛭ icon) > "Run the table" > "On a schedule" > select "Day". 2. Create a conditional run formula for your column that only processes rows where the association wasn't found. You can use an AI formula that checks if the result is empty or has an error. This approach will automatically retry failed associations every 24 hours. While we don't have an exact 12-hour option, the daily schedule is the closest available timing.
Sources: AI Formulas + Conditional Runs [Automated Outbound] - Clay University Datagma integration overview | Documentation | Clay University Scheduled columns | Documentation | Clay University
Loh Y.
·
·
I'm able to do this. But how can i set formula to only run on rows where association wasn't found?
Channeled
APP
·
·
To run your column only on rows where the association wasn't found: 1. Click on your column's "Run Settings" 2. Select "Conditional Run" 3. Click the "Use AI" button 4. Enter something like: "Only run if the [association column] is empty or null" The AI will generate a formula that checks if your association field is blank, allowing the column to only run on rows where the previous attempt failed.
Sources: AI Formulas + Conditional Runs [CRM Enrichment] - Clay University AI Formulas + Conditional Runs [Clay 101] - Clay University Guide: Ways to save Clay credits | Documentation | Clay University
Loh Y.
·
·
Please help. I need to re-run that column after the 1st run and only the association was not found in the first place.
Loh Y.
·
·
For this setting, I'm unsure if there is any implication to re-run rows where the association was already found in the first place and I would prefer for us to not run as I don't want to risk calling too much of the HubSpot api
Loh Y.
·
·
But if you think re-runnign the already successful rows would not cause any problem - do let me know and also your reasoning.
LuisArturo
·
·
Hey there Loh thanks for reaching out, sorry for the long delay here, to clarify currently there is no way to have the schedule column run only run for columns that did not find a results, although this is something that the team is looking into. In this case you could go ahead and rerun this HubSpot integration as there is no credits involved with rerunning this column, as the only thing it would do is just lookup whatever data there is in regard to the Associated company objects
Loh Y.
·
·
Are you saying it's okay to turn this setting on? My question next would be what if an association is already found? Would they try to find the association again?
Loh Y.
·
·
Cause at the point where the row is triggered to find the company association, most of them would just have 1 company associated. But there will be a point in the future where a contact can be associated to 2 or more companies. This don't always happen, but still I'm simply afraid that by re-running the retrieving company association step would cause the past rows to find new company object and then it would re-update the row again - which is not a behavior I want.
LuisArturo
·
·
Sorry for the delay again, if this integration was to run again it would try the find the association again. What you can do here is then is for any integrations down stream from this one you can go and turn off the Auto-update setting inside of the column setup menu
(11 kB)