Preserve custom field values in anonymous leads that are merged with known leads whose fields are empty

[Elliott_Lowe1]

When an anonymous leads is crated during a visit to our website, we set values in ithe newly created record's 'Initial Online Source' field and other fields that tell us the source of their initial visit to our site.  Later, if a known record for the same lead is created by Sales, we don't set the Initial Online Source field and leave it blank.  Then when we send an email to the lead and the lead clicks the tracked link, Marketo merges the anonymous lead  record with the known lead record.  However, the resulting merged record's Initial Online Source field is empty.

This is unexpected as the 'Merging Leads: Behind the Scenes' help article (community.marketo.com/MarketoArticle?id=kA050000000KyszCAC) states that, "When merging a lead with an empty field with another who has a value for that field, the merged lead will have a value in that field."

Paradoxically, Marketo does clear set a standard field's value to empty when automatically merging an anonymous record with a field value and a known record that has no field value.

My suggestion is that when Marketo authomatically merges an anonymous and a known lead record, any field with a value should be preserved if the other recrod's field is empty.

[Anonymous]

Elliott, this sounds like a bug. I'm following up with Engineering about it.

[Anonymous]

Just wanted to circle back with everyone and let you know that this is not a bug. Also, here’s a recap on what happens when an anonymous lead and a known lead are merged.
 

• The following standard fields are passed to the resulting merged lead if the field is empty on the known lead and has a value on the anonymous lead
  • Lead Status
  • Lead Source
  • Original Source Type
  • Inferred fields
  • Anonymous IP
  • Original Referrer
  • Original Search Engine
  • Original Search Phrase
• For the standard Lead Score field, the merged lead keeps the higher lead score
• All activities/memberships of the anonymous and known leads are saved to the resulting merged lead

 
So, Elliott’s idea (which is a good one), is to treat custom fields the same way as described in the first bullet above.

[Dan_Stevens_]

Wow, I did not realize this.  Thanks for surfacing this with the community, Elliott.  Hopefully, this gets "fixed" - even though it's not a bug.  I'm sure most just had assumed this was happening with the anonymous-to-known merging of lead data.

[Anonymous]

Curious how you are doing this. Did you set up smart campaigns to listen for the initial source? If so, are you looking at the original referrer that is populated? 

[Elliott_Lowe1]

Hey Courtney, I dont use the Original Referrer field value because it is only a text field that can hold 255 characters, the original referrer can be much longer, especially on paid ads, so some of the query parameters are truncated and you can't rely on it. Please vote for Change Original Referrer field type from text to text area and maybe Marketo will change this.

Instead, I have a set of smart campaigns that use the Visited Web Page trigger and itsconstraints for each lead source I want to track.  Necessarily these are fairly high level (e.g. Partner, Online Ad, Organic Search, Paid Search, Social Media, Web Referral, Direct Website). If a lead qualifies for a particular smart campaign, their Recent Source field value is set.  If it's their first visit to our website, then another smart campaign sets the Initial Online Source field to the value in the Recent Soruce field .  If it's their first form submission, another smart campaign sets their Opt-In Source field value to their Recent Source field value. When the lead becomes a won opportuniyt, we stop updating the recent source field, so we can see the last online touch   It's a bit tricky, but it provides essential additional insight that we use for allocating our channel spend.
 

[Anonymous]

We just set up similar campaigns I think. We are listening to the value in the referrer url for web page visits. For example if someone comes to our site and the referrer url for the visit contains "quora" we tag that lead coming from quora.  We've set up over 30 campaigns for our top channels - paid, PR, organic search, etc. 

One thing we are struggling with is tagging leads that come in from a random source. For instance if someone gets our link from a blog we don't have a campaign set up to track that lead will get tagged as "direct" because the next wep page the lead clicks on will have our site at the referrer url. 

Do you guys have a system for tagging "other." Would love to hear any ideas. We've been stuck on this issue for a while!

[Elliott_Lowe1]

We have a 'Web Referral' smart campaign for referrals from sources other than the ones I noted in my previous comment.  It has a Visited Web Page trigger has a 'Referrer is not empty' contstraint, a Visited Web Page filter with a Refrrer not starts with [your website domain ] constraint and a series of Member of Smart List filters that indicate it is not in a smart list associated with any other referring source that we track and for which we set a value (e.g. Online Ad, Paid Search, Social Media, etc.).  The smart lists use a Web Page was Visited filter with constraints for the Referrer and Query Paramaeter and 'Date of Acitivy in past 1minute'.

[Anonymous]

Oh man I just implemented this source attribution system based on your recommendations Elliott Lowe and made it fairly more complex with a second set of fields for details and when I started testing what happens in a merge I found this out. Have you found any way around this since then?