Self-Reported Attribution

Self-reported attribution asks the prospect directly: "How did you hear about us?" It captures dark funnel sources that software attribution misses. Podcast mentions, word of mouth, community conversations, social posts seen but not clicked. Software tracks clicks. Self-reported attribution tracks awareness.

The principle: software attribution and self-reported attribution answer different questions. Software tells you which click path led to the form fill. Self-reported tells you what actually made them aware of you. Use both. Neither is complete alone.

Why Self-Reported Attribution Matters

What software misses

Attribution gap

Software attribution typically captures 40-60% of true sources. The remaining 40-60% shows as "direct" or "organic." Self-reported attribution fills that gap, especially for dark funnel sources that drive 50%+ of B2B pipeline.

The Question Design

Primary question

"How did you first hear about us?"

Not "how did you find us" (implies the form fill path, not awareness). Not "where did you learn about us" (implies education, not discovery). "How did you first hear about us" asks about initial awareness.

Answer options

Question design rules

• Single-select, not multi-select. You want the FIRST source, not every source. Multi-select produces data that's impossible to analyze cleanly
• 8-12 options maximum. More than 12 and prospects skip the question. Fewer than 8 and you miss important sources
• Include "Other" with a text field. Every quarter, review "Other" responses. If a new source appears frequently, add it as its own option
• "Colleague or friend" is always an option. Word of mouth is typically 20-40% of B2B pipeline sources. If you don't include it, you can't measure it
• No "N/A" or "Prefer not to say." These are escape hatches that reduce data quality. If the field is optional, a prospect who doesn't want to answer will skip it. If required, they'll pick something

Where to Place the Question

Placement options

Placement rules

• Put it on the demo request form. This is the primary placement. One additional dropdown field adds minimal friction. The data is tied directly to the highest-intent action
• Make it required on the demo form. Optional fields get 40-60% fill rates. Required gets 95%+. The data is too valuable to leave optional. One dropdown takes 3 seconds
• If form friction is a concern, use the thank-you page. Show the question immediately after form submission. "One quick question before your confirmation: how did you first hear about us?" Completion rate drops to 30-50% but zero form friction
• Never rely solely on SDR-collected attribution. SDRs forget to ask. They interpret answers differently. They skip it when the call is going well. Use form-based collection as the primary source. SDR-collected is supplementary detail

Implementation

HubSpot implementation

1. Create a contact property:
   - Name: "How did you hear about us"
   - Field type: Dropdown select
   - Options: [your 8-12 options]
   - Group: Contact information

2. Add to demo request form:
   - Position: last field before submit button
   - Required: Yes
   - Label: "How did you first hear about us?"

3. Create a second property for detail:
   - Name: "Attribution detail"
   - Field type: Single-line text
   - Conditional: shows only when "Podcast", "Community",
     "Conference", or "Other" is selected
   - Label: "Which one?" or "Please specify"

4. Create a report:
   - X-axis: "How did you hear about us"
   - Y-axis: Count of contacts
   - Filter: created in last 90 days
   - Drill-down: by lifecycle stage (MQL, SQL, Won)

Salesforce implementation

1. Create a picklist field on Lead and Contact:
   - API name: How_Did_You_Hear__c
   - Values: [your 8-12 options]

2. Create a text field for detail:
   - API name: Attribution_Detail__c
   - Length: 255

3. Add both fields to the web-to-lead form

4. Map lead field to contact field on lead conversion

5. Create a report:
   - Report type: Leads with converted info
   - Group by: How_Did_You_Hear__c
   - Show: conversion rate, pipeline value per source

Analyzing Self-Reported Data

Primary analysis

Cross-referencing with software attribution

Analysis rules

• Report self-reported and software attribution separately. They answer different questions. Don't merge them into one field. Present both to leadership
• Self-reported is more accurate for dark funnel. Podcasts, word of mouth, communities. Trust self-reported over software for these sources
• Software is more accurate for click-path. Paid search, paid social, email campaigns. Trust software for these. Prospects don't always remember clicking an ad
• Review "Other" responses quarterly. New sources emerge. If "TikTok" or "YouTube" starts appearing in Other, add it as a dedicated option
• Track at the pipeline level, not just the lead level. "30% of leads say podcast" is interesting. "Podcast leads convert to pipeline at 2x the rate of Google leads" is actionable

Common Pitfalls

Data quality issues

Measurement

Pre-Implementation Checklist

• [ ] Question wording is "How did you first hear about us?" (not "find us" or "learn about us")
• [ ] 8-12 answer options covering all major channels including dark funnel
• [ ] "Colleague or friend" (word of mouth) is included
• [ ] "Other (please specify)" with text field is included
• [ ] Single-select, not multi-select
• [ ] Placed on the demo request form (or highest-intent conversion point)
• [ ] Field is required
• [ ] Follow-up text field for podcast name, community name, event name
• [ ] CRM property created and mapped
• [ ] Report built showing source distribution and source-to-pipeline
• [ ] Plan to review "Other" responses quarterly and add new options

Anti-Pattern Check

• Making the field optional. Fill rate drops to 40-60%. You lose half your attribution data. Make it required. One dropdown takes 3 seconds. The friction is negligible
• Using multi-select instead of single-select. Prospect selects "Google search" and "LinkedIn" and "Colleague." Now you have attribution soup. Single-select forces them to identify the FIRST source
• Only tracking at the lead level. "30% of leads say podcast." So what? Track through to pipeline and closed-won. "Podcast leads generate $2M pipeline vs $800K for Google leads" is actionable
• No "Other" option. You have 8 sources in the dropdown. A prospect heard about you on a niche industry podcast that doesn't fit any category. They pick "Google search" as a default. You lose the real attribution. Always include "Other" with a text field
• Never updating options. Your dropdown has been the same for 18 months. Meanwhile, you started a YouTube channel, launched a community, and appeared on 12 podcasts. Update options quarterly based on "Other" responses and new channel investments
• Replacing software attribution with self-reported. They answer different questions. Software tracks click paths. Self-reported tracks awareness. Use both. Report them separately. Let leadership see both views
• Asking on a post-meeting email survey. 10-15% response rate. The prospect has already forgotten where they first heard about you. Memory bias is worst at this stage. Ask at the moment of highest intent: the form fill