Blog | Pedal Lucid

What Exactly is 'Source,' Anyway?

Written by Duncan McGovern | Jun 30, 2021 4:00:00 AM

Part 2 of a series about marketing attribution and analytics
Link to Part 1: Thinking Critically About Lead Source

Introduction

Part 1 discussed some of the basics behind using a ‘lead source’ field to get useful information: that it shouldn’t change, how to define and track conversion metrics, and distinguishing between the source of a person and the source of an opportunity.

This second installment covers different ways to categorize ‘source.’

‘Source’ By Any Other Name

There are a variety of ways that source can be tracked; here are some distinctions that we think are worth paying attention to.

Person vs Account

The value of account-level metrics differs from organization to organization, but we think looking at Account source as well as person source is useful for many of our clients. Specifically, it provides a reporting mechanism for determining the original point of awareness/engagement for the entity making a purchasing decision (the single account).

Example:

If a paid digital campaign brings in a net new account, it’s common for the person sources at that account to look something like (in chronological order):

  1. Paid digital campaign - first person who engaged at the prospect account is linked to the ad campaign

  2. Inbound ‘contact us’ request - a different colleague who reached out to follow up with a question

  3. Outbound - another person included on email thread & added to database manually

Looking at person acquisition it looks as though three channels each sourced one potential client; the real story is a digital campaign creating interest at the Account which drives internal interest across multiple subsequent channels. Capturing source per person can still be a useful standalone metric and is a requirement for capturing Account source consistently (which is determined based on the first contact associated with the Account). That said, tracking first touch at the Account level will provide better reporting on how marketing efforts (paid digital ad) create pipeline than strictly using a per-person reporting strategy.

Channel & Campaign

The simplest way to track source is via the biggest useful bucket you have—marketing channels. These might include:

  • Inbound/Unknown
  • Organic Social Media
  • Digital Ad
  • Event
  • Referral/Word of Mouth
  • ...Etc

These channels don’t change month-to-month or year-to-year; your CRM might have a default set of channels available out of the box as a starting point.

The next level of detail is the specific campaign source. This is generally (although not always) timebound, and best thought of as a specific initiative. ThesecCampaigns look like:

  • A specific event
  • A digital ad campaign on a specific platform
  • A referral program with a specific partner

Note that these examples are channel-specific; it’s important to be able to differentiate between the overall initiative (campaign) and channels within that campaign. Different tools approach this in different ways. Hubspot uses source / drilldown 1 / drilldown 2; Salesforce uses campaign hierarchy to accomplish a similar result.

It can also be helpful to store inbound/unknown touchpoints in a similar format to specific marketing initiatives. We may not know exactly what drove someone to sign up for your email newsletter, but a standardized structure lets us ask questions like how inbound newsletter signups convert compared with paid ads.

“How Did You Hear About Us?”

The usefulness of this question starts to break down for organizations with long, multi-touch relationships with their clients—i.e., the kind of organization we typically work with! It assumes anyone filling out the form has not interacted with you before and causes reporting issues by either overwriting existing source data or creating discrepancies between what you’re tracking and what they said. That said, asking people what drove them to your organization is sometimes the only way to track something that’s important—earned media such as a feature in a prominent national publication, for example.

The two-pronged approach we suggest for this dilemma:

  1. When asking ‘How did you hear about us’, be careful to only do so in the context that is limited to new prospects, such as a newsletter signup. A better question on something like a donation form (mix of new & returning donors) is “What inspired you to give today?”

  2. Store answers in a separate, dedicated field such as “Self-reported source” to avoid corrupting existing data. Use automation and/or periodic manual review to determine how you’d like to categorize responses that match your standard source reporting strategy. This leaves the door open to collecting useful data while storing it in a separate place to keep source data quality consistent.

UTMs

UTM Parameters are ways to add tracking capabilities to a website url, short for “Urchin Tracking Module” which is a product acquired by Google many years ago. They appear in the url and use the following format: pedallucid.com?utm_source=facebook&utm_medium=organic_social. This example will take you to the Pedal Lucid homepage and indicate the click was sent there from ‘facebook’ (source) via an ‘organic_social’ click (medium). UTMs are used by Google Analytics and other marketing tools to tie website visitors to specific marketing efforts using the following parameters:

  • Medium
  • Source
  • Campaign
  • Term
  • Content

Google Analytics as well as marketing automation platforms such as Hubspot or Marketing Cloud are able to parse these values and capture them in your database for engagement and attribution reporting. Depending on the tool you use, utm values are parsed in slightly different ways. The important thing to consider is how to ensure values are standardized in a way that provides consistent reporting. For example, which “source” field in your CRM aligns with “utm_source” (platform) and which to “utm_medium” (channel)?

Putting It All Together

You aren’t limited to a single ‘source’ datapoint; it’s certainly possible to capture all UTM fields, self-reported source, and other custom app-specific values. That said, we recommend starting with the two-tiered approach of channel & Campaign. Once you’ve nailed these two, you’ll be well-positioned to decide if and how to incorporate additional data points.

In this sample report below, we can see which marketing channels brought in the greatest number of leads (Purchase List) as well as how each specific Campaign within that channel performed using a stacked bar chart. Adding additional variables is certainly possible, but makes the reporting multi-dimensional and harder to view at a glance what’s performing well.

Note that this chart displays simply the sum of total Leads created, but could be reconfigured to display other metrics such as total converted Leads or conversion rate. 

Each organization has a unique set of needs that will shape how the channels are defined. Some organizations may be fine with a single ‘Event’ category. Others might want to distinguish between external events (tradeshow, tabling, etc) and internal events your organization hosts. If you have a very specific type of event that would be helpful to see as a standalone category (say, group volunteer days), you might choose to break that out as its own category. The key is to maintain consistency in the framework you land on.

At a tactical/implementation level, we categorize Salesforce Campaigns by marketing channel—either with the standard ‘Type’ field or a custom picklist. The first time a Lead or Contact is added to a Campaign, we’ll use Flow to populate custom ‘Campaign Source’ and ‘Channel Source’ fields on that record. Once populated, those fields won’t be overwritten and become your source of truth for acquisition metrics. As long as people are being correctly added to Campaigns (either by staff or a marketing automation platform), Salesforce automatically takes care of tracking the first touch/sourcing point of engagement.

Up Next

In Part 3, we’ll talk about the basics of attribution models. You may also be interested in Pretty Good Answers, a separate post about what to track in your CRM vs website analytics tool.