You Probably Don't Know Where Your Users Actually Come From

There's a question every founder or growth lead asks eventually, usually after spending a meaningful chunk of budget on podcasts, YouTube, or influencer campaigns:

Which of these is actually working?

The uncomfortable truth is that most teams can't answer it. Not because they haven't tried — they've set up UTM parameters, they've got Google Analytics, they've given creators unique links. But the data they're getting back is incomplete in ways they don't fully understand, and they're making decisions based on partial information.

This guide is about fixing that. Not with a complex technical stack, but with four specific signals that together give you a complete picture of where your customers are actually coming from.

---

Why "just use UTMs" isn't enough

UTM parameters are valuable. If you're running any kind of digital advertising, you should absolutely be tagging your URLs. But UTMs have a fundamental limitation for creator advertising: they only work when someone clicks the tagged URL.

For podcast, YouTube, and creator content, the conversion path is frequently indirect:

• The listener hears an ad and visits the website directly hours later
• The viewer sees a custom URL overlaid on screen and types it manually
• The customer hears a promo code and uses it at checkout three days after listening
• The subscriber remembers the brand name, Googles it, and converts organically

In all of these cases, your UTM attribution either misses the conversion entirely or assigns it to "direct" or "organic" — which tells you nothing about which creator drove it.

---

The four signals you need

Signal 1: Tracking links

The most familiar signal. Give each creator a unique tracking URL — something like yoursite.com/go/james — and count how many clicks come through it.

What it captures: Listeners who click the link directly, typically from show notes, YouTube description, or swipe-up links.

What it misses: Everyone who doesn't click the link in the moment — which, for podcast advertising, is often the majority of converters.

How to use it: Track clicks and conversions through the link. Use a platform like Castlytics that ties the link to a specific campaign and creator, so you can see attributed revenue, not just click volume.

Signal 2: Vanity paths

A vanity path is a short, memorable URL extension that maps to your product — something like yoursite.com/james or yoursite.com/theleveragepod.

When a podcast host says "go to yoursite.com/james to get started", their audience is far more likely to type that URL than to hunt for a show notes link. The vanity path captures those visits as a distinct attribution signal, separately from tracking link clicks.

What it captures: Direct URL visits from listeners who remembered and typed the vanity path.

What it misses: Conversions that happen without any URL visit — promo-code-only checkouts, for example.

How to use it: Create a vanity path for every creator you work with. Castlytics tracks vanity path visits as campaign_landing events tied to the campaign, so they appear in your attribution data alongside link clicks.

Signal 3: Promo codes

When a creator reads a promo code on air ("use code JAMES for 20% off"), a subset of their audience will use that code at checkout — without ever visiting a tracking link or vanity path.

Promo code matching captures these conversions. By tying a specific promo code to a specific campaign, every checkout that includes that code gets attributed to the creator who read it.

What it captures: Conversion-intent listeners who went directly to your site or app and used the code at checkout.

What it misses: Visitors who converted without using a code, and those who used a generic discount code not tied to the campaign.

How to use it: Create a unique promo code for each creator. Pass order data including promo codes to your attribution platform so they can be matched to campaigns.

Signal 4: Post-purchase surveys

The simplest signal, and often the most revealing. Ask every customer at checkout or in a post-purchase email: "How did you hear about us?"

When a customer types "The Leverage Podcast" or "heard it on James's show", you've got a direct attribution signal — one that required no link clicking, no vanity path visit, and no promo code. Just a customer telling you the truth.

What it captures: The dark funnel — customers who converted through channels that digital tracking can't see at all.

What it misses: Customers who don't fill in the survey, and those who give vague answers ("online", "a friend").

How to use it: Add a single open-text question to your checkout or post-purchase email. In Castlytics, survey responses are automatically matched to campaigns by creator name, handle, or podcast name — no manual matching required.

---

Putting the signals together

Here's what multi-signal attribution looks like in practice.

Imagine you're running three simultaneous campaigns:

• The Leverage Podcast — Tracking link + vanity path yoursite.com/leverage + promo code LEVERAGE20
• James Horwath YouTube — Tracking link + vanity path yoursite.com/james
• The Creator Brief newsletter — Tracking link + promo code CREATOR

After 30 days, your data might look like this:

| Campaign | Link clicks | Vanity visits | Promo redemptions | Survey mentions | Total conversions | |---|---|---|---|---|---| | The Leverage Podcast | 48 | 31 | 19 | 12 | 94 | | James Horwath | 61 | 22 | — | 8 | 73 | | The Creator Brief | 29 | — | 14 | 4 | 39 |

Without the vanity path, promo code, and survey signals, The Leverage Podcast appears to have generated only 48 conversions. With all four signals, it's 94 — nearly double. Your budget decision for next quarter looks very different.

---

Attribution models: first, last, and linear

Once you have multiple touchpoints for a single conversion, you need to decide how to attribute it.

First-touch attribution gives 100% credit to the first campaign a visitor encountered. Useful for understanding brand awareness and top-of-funnel impact.

Last-touch attribution gives 100% credit to the final campaign touchpoint before conversion. Useful for understanding what pushed someone over the line.

Linear attribution splits credit equally across all touchpoints. Useful for a balanced view when customers interact with multiple campaigns before converting.

None of these is definitively correct. The right model depends on your marketing mix and what question you're trying to answer. The important thing is to use a tool that lets you switch between them so you can see the picture from multiple angles.

---

Setting this up without an engineering team

The four-signal model sounds complex, but in practice, getting it running doesn't require a developer.

1. Create campaigns in Castlytics — one per creator, with their tracking link, vanity path, and promo code
2. Share the tracking link with each creator for show notes and description boxes
3. Configure the vanity path — a simple redirect rule on your domain, or handled automatically by Castlytics
4. Pass promo code data — via Castlytics's JavaScript tracker or the REST API
5. Add a survey question — one line of copy in your post-purchase email or checkout confirmation page

That's the full setup. The attribution engine runs automatically from there.

---

The bottom line

If your current answer to "where are your users coming from?" is "mostly direct and some organic, plus some clicks on the podcast links we sent out" — you're missing most of the picture.

The actual answer requires four signals working together. The good news is that setting them up is straightforward, and once they're running, you'll never have to guess which creator is worth your budget again.