Data on Matt Lincoln - Salesforce, Pardot and Martech.

Why isn’t a field being returned in the Marketing Cloud Account Engagement (Pardot) API request?

Fri, 24 Jun 2022 00:00:00 +0000

Today I encountered an interesting Pardot API question I’d never seen before, which was worth some exploration: A developer I’m working with was trying to manipulate a Pardot field - however when he performed a GET request on a Prospect to ensure the existing field value could be found, the field was missing from the API response.

Investigation

To get down to the solution, I like to use Postman for API testing. After following the API Authentication documentation, I connected to our lower environment (UAT) and pulled an example record’s details with the Prospect Read operation using a GET request in the following format:

<https://pi.demo.pardot.com/api/prospect/version/4/do/read/email/a@b.com&format=json>

When the response is returned, I can see all the details I would expect, except for that my desired field, Corporation ID, is not returned in the response:

As this is unexpected, we check the Pardot org. Are we perhaps logging into the wrong org/business unit? Did we create the custom field in a different org and not this one?

After verifying that everything is correct, we’re logging into the right business unit and that the field does indeed exist, we check a few other fields. As we can see from the first screenshot, we expect to see a response returned for all fields, even if the values are null… or do we?

I can see the other custom fields which are populated, are returned in the API response, however all the custom fields which are not populated, are not returned in the response.

So if we add a value to the field we’re looking for, we can test the hypothesis that custom fields are only returned if not equal to null.

After running our GET request again, we see that now, the field and value are both provided in the API response:

Summary

We can see that custom fields with null values are not included in the Pardot API response, whereas custom fields with values are included. On the other hand, standard fields are always included, regardless of whether they are valued or null.

	Valued	Null
Standard Fields	Included in API Response	Included in API Response
Custom Fields	Included in API Response	Not included

Contacts to Multiple Accounts and Pardot

Mon, 01 Nov 2021 00:00:00 +0000

Introduction

The first question you’ve probably got is what’s the image at the top of the page all about?

Whenever I think about Contacts to Multiple Accounts, I’m always reminded of the phrase, “wearing more than one hat”, which means having more than one role or responsibility at the same time.

That’s the scenario C2MA was designed to support - when one person has an association to more than one company or organisation.

I’ve transcribed my talk from Pardreamin 2021 below, for anyone who wasn’t able to make the session. Here’s a link to the slides.

Why is Contacts to Multiple Accounts needed?

Let’s start with looking at a small bit of theory to give some context to anyone unfamiliar with Contacts to Multiple Accounts.

I’ve got really tired of writing that in full over the last year or so, so from here on in, you might see it written as C2MA instead.

It’s useful to start by looking at the default Salesforce object model, and checking how Accounts and Contacts relate to each other.

As you can see from the diagram, an Account can have many Contacts related to it, but a Contact can only be related to one Account out of the box.

So this is a problem if you’re dealing with decision makers who “wear more than one hat”.

I’ve got a few examples here, some of the most common ones I’ve seen in person or read about.

Maybe your target market is the healthcare industry… it’s quite common for clinical practitioners such as doctors, dentists and opticians to work for more than one hospital or healthcare facility
Or perhaps you’re targeting marketing consultants, who might have a vendor relationship to multiple accounts as an external consultant.
Also don’t forget about business owners or entrepreneurs, who might have founded or own more than one company. It’s worth thinking about How would you represent Elon Musk as a Contact in Salesforce, if you were trying to sell to SpaceX and Tesla at the same time?
Beyond that, a couple more examples could be that a person can join more than one membership organisation, or donate to multiple charities

How to enable Contacts to Multiple Accounts

Feature Settings → Sales → Account Settings → Edit → Allow users to relate a contact to multiple accounts

I’ve included a link here to the implementation guide, and if you follow that, one of the steps you’d take is to add your Related Accounts List to the Related List section of your Contacts Page Layout, and this Related Contacts List to your Account Page Layout.

So ultimately you’ll be able to see which Contacts are associated with each Account, and which Accounts each Contact is associated to.

Couple of nice benefits to mention are that you can set whether the Contact is still Active at an Account, which solves that question everyone always has around “What should I do when a Contact leaves the company?” You can also set the start and end dates of them being associated to the Account - probably overkill to be honest.

Another thing I like is that you can set the Role the Contact has at each Account - it’s a clean way to define who are the gatekeepers, decision makers and influencers.

The Direct field is also really important. You can’t edit it directly from the Account Contact Relationship record. The Direct checkbox gets ticked when that account is set as the Contact’s Primary Account, which means it’s the Account selected under the Account Name lookup field on the Contact record. Any other accounts added through the Add Relationship button on that related list, become non-direct Account links, where the link between the Contact and the Account only exists through the Account Contact Relationship object.

Watch out for Account Contact Roles

There’s also a separate older feature called Account Contact Roles, which is very easy to confuse with Contacts to Multiple Accounts, but in fact is a completely separate feature. Two key points why it’s less attractive for us to use

Firstly it’s only available in Salesforce Classic and not Lightning (the implication being that it’s no longer seen as part of Salesforce’s roadmap)
Second, it’s not as attractive because you can’t create custom fields, or use validation rules or process builder on it

So what’s going on behind the scenes?

Before we enable Contacts to Multiple Accounts, if we want to represent a Contact’s relationship to a Parent Account and two Child Accounts, and we want to be able to send personalised emails about data we hold about each of those Account records, we need to duplicate the Contact multiple times and have one Contact linked to each Account. I don’t need to go into too much detail on why that’s a problem, but at a high level, you’re not getting a single view of that person’s interactions with you, and it makes things like marketing preference management challenging.

After enabling contacts to multiple accounts, we get the scenario in the image above, where a single Contact can have that direct relationship to an account, and then we can have records on the Account Contact Relationship junction object which sits between the Account and the Contact, and that forms an indirect relationship between the Contact and any additional Accounts.

The summary is that we can segment our audience based on data on the non primary Account records associated to a contact - we’re just not going to be able to personalise those emails.

What does it mean for Pardot?

Pardot’s designed to take a Prospect, typically your B2B decision maker, on a Sales cycle, from awareness, through to sale and then help retain them as a client.

You can consider the Prospect as a single entity, which as a marketer, you combine with data from the Related Account and Related Opportunities, and you’re bringing that decision maker along your sales cycle by targeting them with personalised and relevant comms.

But if you’re a business where each Account associated with a prospect is considered an entirely separate entity with a different Sales cycle and needs engagement and personalisation specifically on the Account level, this is where the Pardot data model falls short.

The main problem you’ve got is that we can only personalise emails based on the Account related to the Salesforce Contact with the Lookup field, which is what we called the Direct relationship earlier. It means we can never bring in things like the Account name, or any relevant personalisation details such as renewal date, address details, or any products or services the account has purchased.

So the question is, if we can’t do what we want out of the box, what can we do?

If we want to segment our audience based on fields on those non primary Accounts, we need to use Custom Objects in Pardot to map our Account Contact Relationship object.

We can then build Dynamic Lists in Pardot to build lists of Prospects who match certain criteria such as “Show me a list of Prospects associated with an Account with a subscription date which expires in the next 30 days”.

The challenge here is that by default the Account Contact Relationship record won’t hold information from the related Account, so you’ll need to use a Salesforce-side automation such as Flow to pass down information from the Account record to the Account Contact Relationship record related to it. You can then consume that data in Pardot.

That leads to the next point, where another workaround is that Pardot might not be the right tool for after you initially onboard a client, or it might not be the right tool for you at all.

Just a brief mention that Marketing Cloud has the capabilities to build a custom data model with its Data Extension functionality, so you’re not tied to the Prospect data model we looked at in Pardot. It’s a higher price solution, but really you’ve got more control over your data structure and email personalisation logic. Salesforce themselves do recommend using Pardot and Marketing Cloud together in some scenarios where you need higher levels of personalisation after the initial sale to a client.

What we learned from a crowdsourced Pardot Health Check Audit of over 100 accounts

Sun, 30 May 2021 00:00:00 +0000

Introduction

If you manage one or multiple Pardot business units, you’re probably well aware of the importance of keeping your accounts in good shape. Since account maintenance isn’t always the most glamorous part of your role, you may end up putting it off for a rainy day. However, this can often have consequences, such as a poorly-optimised campaign, or a lack of insight into key metrics.

One of the most important steps to maintaining a successful marketing technology platform is making sure that your process is set up optimally for the way your organisation works. It takes time to set up your Pardot account, which is why it’s important to know how to do it, and what ‘gotchas’ to look out for.

At Pardreamin’ 2020, I wanted to give some best practice tips on doing a Pardot health assessment. However it was a great opportunity to make a hands-on session, and allow participants to not only see their own org’s health in isolation, but also to compare their org against their peers in the first crowdsourced Pardot health check. I’m often asked “what’s the average of this or that metric?” by clients, but this was a chance to get some concrete data for comparison.

Want to participate yourself?

Follow the walkthrough guide on The DRIP.

Watch the session recording:

The session took the following structure:

Average % of Mailable Prospects = 78.4%

How did we measure this?

This metric was calculated by dividing the total number of Prospects who can receive emails, e.g. not marked as Opted Out, Bounced or Do Not Email, by the total number of Prospects, then calculating the average across each org.

What can we learn from this?

The metric tells us that on average, around 1 in 4 Pardot Prospects are not eligible to receive emails.

It’s a fairly high number, and tells us that perhaps Pardot Admins aren’t spring cleaning their account regularly. It’s worth asking yourself if your marketing team has a good reason to be holding data in Pardot which you can’t send emails to.

The interesting thing is that the actual number of mailable Prospects will be even lower than this in the real world. The limitation in this measurement is that for clients using Email Preference Pages, prospects can opt out of specific communications without being counted as a globally opted out Prospect in your account. Only accounts exclusively using the “global unsubscribe” option get an accurate number of “opted out” Prospects.

What can I do if I score poorly (low percentage)?

It’s not necessarily a negative - unmailable Prospects don’t count towards your database limit. Consider whether the prospects are Opted Out or set as Do Not Email, or both. They’re all different scenarios and require different approaches:

Opted Out and Do Not Email:

Clicks the unsubscribe link in a Pardot email or email preference page.
Reports an email as spam.

Do Not Email:

Has a hard bounce or five soft bounces.

Opted Out:

Is imported as opted out.
Is opted out in Salesforce.
Is manually opted out in Pardot.

Next Steps

Reading the Pardot Database Cleaning Guide
Consider archiving Prospects if they don’t want to receive comms from you, or you’ve already imported as “Import prospects and global opt-out”
Enabling Prospect Resubscribe
Cleansing any “bounced” emails - expose these Leads or Contacts to reps in Salesforce List Views by using the Pardot Hard Bounced field or enabling Salesforce Bounce Management.

Mailable Prospects who have been emailed? - 76.1%

How did we measure this?

This metric was calculated by taking the number of mailable prospects and dividing it by the number of prospects who have ever been sent an email to in your Pardot org.

What can we learn from this?

The aim of this metric is to understand wasted database capacity.

If around 25% of mailable Pardot Prospects are never sent emails, it means that Pardot customers are generally overpaying for their license subscription.

If we do some really rough maths, our experiment gives an average of 79,169 mailable Prospects across the participating accounts, and the average account has emailed 50,114 mailable Prospects. This equates to a 29055 average “overage” in unneeded Prospect capacity.

The latest US pricing PDF shows that at the minimum pricing level, 10,000 additional Prospects are priced at $1,200 per year. This means that the average client is paying for 2.9 “blocks” which it doesn’t need (29055/10000).

Salesforce tells us that the Plus tier is the most popular, so if we extrapolate out $1,800 x 2.9 = this means an average of $5,220 per year in unnecessary spend for each account participating in the benchmark.

What can I do if I score poorly (low percentage)?

The first thing to check is whether you score low because you use an Email Preference Page. There is a limitation that clients using an Email Preference Page may have Prospects who are mailable from a technical perspective, but unmailable from a permissions perspective.

If not, it’s time to start thinking about why you hold prospect data which you don’t send emails to.

Consider why are these Prospects in your database and should they still be there? The Dynamic List created as part of the experiment brought back a list of Prospects who HAD been emailed, but let’s create a new Dynamic List to fetch the records which HAVEN’T, by filtering for Prospects who are not on the list of mailable Prospects who have received an email.

Take a look at the Prospects which enter this Dynamic List. Work out how they entered your database by looking at the Campaign, Conversion Point, or the audit tab.

Are your prospects missing key segmentation fields? Alongside checking individual Prospect records for missing data, take a look at the last 10 or 20 emails sent from your Pardot account across List Emails and Engagement Studio start lists. Whilst using Dynamic Lists in Pardot for segmentation purposes is generally best practice, you might not realise that large parts of your database just don’t get picked up due to the filter criteria you use.

Prospects never active, or not active in the last 6 months? - 70.3%

How did we measure this?

A prospect is known as active when they interact with our Pardot assets, such as clicking a link, viewing a page or submitting a form. However, email opens are not counted as many spam filtering applications open emails to check the content. It’s easy to measure this using the Prospect time filter on Dynamic Lists. In the audit,

What can we learn from this?

This metric is a great indicator of understanding how engaged your database has been recently with your Pardot marketing activities. We can see that 70.3% of prospects have never been active on average, meaning that the vast majority of a Pardot org’s Prospects aren’t interested in the emails you’re sending them.

To the marketer clicking the send button, every email they’re churning out feels important, but sadly that’s rarely the case for the recipient. This metric highlights the importance of considering every member of your database as an individual, and targeting them as specifically as possible based on the segmentation information you hold about them in your database.

What can I do if I score poorly (high percentage)?

The measurement of a Prospect as being ‘active’ revolves around the concept of ‘conversion’ in your Pardot org. Conversion is the link between a cookied ‘visitor’ on your website and Prospect in your database. To measure a Prospect being active, they must have converted. It’s worth thinking about how you elicit prospects to complete your webforms and click on CTA buttons in your email templates - issues with these normally goes hand in hand with low conversion rates.

The other common reason for a poor performance for this metric is if you’re not actually sending emails to your database, it’s only natural that you’re consequently going to have a lower level of activity. It’s worth combining this Dynamic List with the previous one, to get a feel for which Prospects aren’t engaging with your marketing assets because you’ve not been emailing them, versus who isn’t engaging because they just aren’t interested.

Check email click through rates overall. Do you use A/B testing to see what makes your audience ’tick’? SPF authentication? Email CTAs?

What’s the source of the Prospects? Are they not interested in your marketing efforts because you’re targeting the wrong audience? Or, if you’re definitely targeting the right audience, can you segment them more finely based on what you know about them to improve engagement?

If all else fails, should these Prospects or a subset of them be “spring cleaned” and put into the recycle bin, in order to tidy up your account and keep within your mailable Prospect limit? Before you do this, is a re-engagement campaign appropriate?

First Name Missing? - 5.0%

How did we measure this?

This Dynamic List used the “Prospect field “is empty” criteria. We divided the number of Prospects with a missing first name, by the total number of Prospects in each account.

What can we learn from this?

The point here wasn’t really about understanding the level of which the first name field is populated. That’s one small data point amongst many fields used for personalising emails.

Even so, the number is surprisingly high. That’s 1 in 20 records on average which is missing a first name. This is typically seen alongside what I call a “Mailchimp mentality” of just uploading email addresses for a particular unpersonalised “email blast”. One of the biggest mentality shifts for a new client is that you should consider Pardot not as a tool to send individual emails, but instead think of it as a platform to build a long term Prospect Database.

What can I do if I score poorly (high percentage)?

The most successful Pardot users have a defined set of data points required before a Lead can be passed across from Marketing to Sales. If you haven’t already, define your MQL (Marketing Qualified Lead) criteria. This ensures that Sales receive Leads which are well qualified, consistent in quality and are more likely to convert. Could a Sales rep work a Lead which doesn’t have a name? Every sales process is different, but it’s likely for most organisations that the answer is no. Audit all the data collection points where Prospects enter your Pardot database. Whether it’s through Pardot forms, imports, Connectors, or the API, ensure that you’re not missing the chance to collect data in fields which are required for your Sales process - or put in a plan to collect them using Pardot functionality such as Progressive Profiling.

Ask yourself, for Prospects missing key data points, are those records really good quality data? Typically it can be the sign of a bad quality dataset being imported into your account. Do you really have permission to hold this data in your org? Consider sending these records to the recycle bin until they submit a form and provide complete information. If you don’t even hold a first name for a Prospect, how can you be sure you know enough information about them to know which types of emails you should be sending them?

Also consider exposing these records to your Sales team using List Views in Sales Cloud. Perhaps they can populate the missing information (were the records synced from Salesforce in the first place and reps didn’t complete the first name or any other field because it wasn’t mandatory? Perhaps alternatively, a data provider could enrich your database with missing infromation?

Have a backup plan for blank values used as merge fields:

Ensure a default merge value is set in case of a blank field value
Use HML conditional statements to have more fine-tune control, for example:

    {{#if Recipient.FirstName}}
    Hi {{Recipient.FirstName}},
    {{else}}
    Hello!
    {{/if}}

Apply this approach to check other fields, e.g….

Last Name
Job Title
City
Country
Segmentation
Industry
Company Size (Employees/Turnover)

What could have been better?

It wasn’t a perfect experiment. There were some nonsensical submissions which I’ve removed from the data set, for example, submissions where the number of Prospects which had never received an email were higher than the total number of Prospects in the database! If I were to do this Pardot health check again, I would:

Encourage participants to take numbers from the email showing that a Dynamic List has completed refreshing, as I suspect some participants didn’t give their Dynamic Lists enough time to refresh
Should have showed participants how to clear any existing filters in the Prospect table before checking the total number of prospects
Based on the comments and questions I’ve received, I’d encourage participants to not take the Dynamic Lists I shared too literally - e.g. checking blanks in the first name field is just an introductory concept - apply it to your other fields!
If I were to run the session again, I’d try to squeeze in the following health check points:
- Prospects not synced to Salesforce - there’s less of a “correct” answer here, but it’s an important metric to evaluate the qualification process used across Pardot and Salesforce.
- Prospects who have been overemailed - this can be done easily using the “Prospect has been emailed at least X times in the last X days” criteria and is particularly important in accounts with high numbers of different users with permissions to send emails.

In summary, it’s best practice to regularly run a Pardot health-check auditon your account to make sure you’re getting the most from the system, and that yourmarketingchannels are giving you the best possible return on your investment.

Methodology Note

There were two options for methodolgy here when calculating the averages used below:

a) Calculating statistics on a prospect level across Pardot accounts, e.g. creating a total sum of Prospects across all orgs, then using this as a denominator for each calculation, e.g. with total sum of mailable Prospects as a numerator.

b) Calculating an average of each Pardot account’s metrics, e.g. finding the percentage of mailable Prospects for each Pardot account individually, then calculating an average of each metric across all accounts

Option B was selected as a better fit as it was less impacted by org-level outliers, e.g. the potential for a huge Pardot org with 2 million prospects to wildly skew the average number of prospects and consequently the averages for each of the metrics covered in this article.

Data on Matt Lincoln - Salesforce, Pardot and Martech.

Why isn’t a field being returned in the Marketing Cloud Account Engagement (Pardot) API request?

Investigation

Summary

Contacts to Multiple Accounts and Pardot

Introduction

Why is Contacts to Multiple Accounts needed?

How to enable Contacts to Multiple Accounts

Watch out for Account Contact Roles

What does it mean for Pardot?

Further Reading

What we learned from a crowdsourced Pardot Health Check Audit of over 100 accounts

Introduction

Average % of Mailable Prospects = 78.4%

How did we measure this?

What can we learn from this?

What can I do if I score poorly (low percentage)?

Next Steps

Mailable Prospects who have been emailed? - 76.1%

How did we measure this?

What can we learn from this?

What can I do if I score poorly (low percentage)?

Prospects never active, or not active in the last 6 months? - 70.3%

How did we measure this?

What can we learn from this?

What can I do if I score poorly (high percentage)?

First Name Missing? - 5.0%

How did we measure this?

What can we learn from this?

What can I do if I score poorly (high percentage)?

What could have been better?

Methodology Note

Further Reading