🌌 Path to Conversion by Campaign Type

Dashboard link: https://app.emax-digital.com/amc-path-to-conversion-from-last-campaign

Overview

The Path to Conversion by Campaign Type dashboard uses Amazon Marketing Cloud (AMC) data to reconstruct the sequence of campaign types a shopper was exposed to before they purchased. Instead of crediting only the last ad clicked (the default Amazon attribution model), this dashboard maps the full journey — up to five touchpoints — and shows which combinations of campaign types actually drive conversions. It also lets you compare first-touch vs last-touch attribution, so you can see which campaign types open journeys (awareness role) versus close them (conversion role).

This is one of the most useful dashboards for answering: "Is my upper-funnel spend (DSP, Sponsored Brands) actually contributing to sales, or is Sponsored Products doing all the work?"

This dashboard requires an active Amazon Marketing Cloud (AMC) integration.

What You Can See on This Dashboard

Section 1 — Unique Users by Path Step

The top of the dashboard shows how many unique users were exposed at each step of the conversion path, broken down by campaign type at that step.

Key metrics:

• Unique Users per Step — Count of distinct shoppers reached at each step (1 through 5) of the conversion path.

• Campaign Type at Step — Which campaign type (DSP, Sponsored Products, Sponsored Brands, Sponsored Display, etc.) was the touchpoint at that step.

Visualizations:

• Bar chart with Step (1–5) on one axis and Unique User count on the other, broken down by campaign type at that step.

How to read it: Step 5 typically contains the most users — this is by design. All paths terminate at step 5, so shorter journeys (1, 2, or 3 touchpoints) are still represented at step 5 as their final converting touchpoint. The earlier steps show which campaign types are pulling shoppers into longer journeys.

Section 2 — Path to Conversion by Campaign Type (Detail Table)

The main analytical table of the dashboard. Each row is a unique sequence of campaign types that led to a conversion, with the full economics attached.

Columns shown: Ct 1 · Ct 2 · Ct 3 · Ct 4 · Ct 5 · Last Campaign · Path Occurrences · Impressions · Purchases · Ad Spend · Total Purchases · Sales · Total Sales · ROAS · User Purchase Rate.

Key metrics:

• Ct 1 – Ct 5 — The campaign type at each of the five touchpoints in the path (some steps may be empty for shorter paths).

• Last Campaign — The final campaign type the user was exposed to before converting.

• Path Occurrences — How many users followed this exact path.

• Impressions — Total impressions delivered across all five touchpoints in this path.

• Purchases — Ad-attributed purchases from the Last Campaign only.

• Ad Spend — Total ad spend across all five touchpoints in the path (not just the closer).

• Total Purchases — Total purchases (ad-attributed plus organic) associated with this path.

• Sales — Ad-attributed sales from the Last Campaign.

• Total Sales — Total sales (ad-attributed plus organic).

• ROAS (Return on Ad Spend) — Total Sales ÷ Ad Spend across the full path.

• User Purchase Rate — Share of users on this path who converted.

Important attribution note: Ad Spend is summed across all touchpoints of the path, while Purchases / Sales are credited to the Last Campaign only. ROAS here therefore answers: "For every euro spent across this full journey, how much revenue came back via the last touch?" This is a more honest measure of cross-channel efficiency than a single-campaign ROAS.

Section 3 — First versus Last Touch View

A reframing of the same data through an attribution lens: instead of looking at the full sequence, every path is collapsed to its first touchpoint and its last touchpoint, and the economics are aggregated.

Columns shown: First Touch · Last Touch · Ad Spend · Impressions · Total Purchases · Total Sales · Total ROAS.

Key metrics:

• First Touch — The campaign type that opened the journey (step 1).

• Last Touch — The campaign type that closed the journey (the converting touchpoint).

• Ad Spend / Impressions / Total Purchases / Total Sales / Total ROAS — Aggregated for every path matching that First-touch → Last-touch combination.

How to read it: This table shows you the difference between demand creators (campaign types that frequently appear as First Touch) and demand closers (campaign types that frequently appear as Last Touch). DSP and Sponsored Brands often skew toward First Touch; Sponsored Products often skews toward Last Touch.

Available Filters

Filter interaction notes

• Report Date is required. The dashboard will not render until a date is selected.

• Last Campaign is dynamic — the dropdown only shows campaigns that actually closed at least one path in the selected period and marketplace.

• Path Includes vs Last Campaign: Path Includes asks "did this campaign type appear anywhere in the journey?", while Last Campaign asks "was this campaign the final touchpoint?". Use them together to isolate, for example, paths where DSP appeared somewhere and Sponsored Products closed the deal.

• Show redacted = Yes will surface paths Amazon has anonymised due to small audience sizes. These rows have limited diagnostic value but help reconcile totals.

Use Cases

1. Prove the value of upper-funnel spend — Open the First versus Last Touch table and sort by First Touch = DSP. Look at the Total Sales and Total ROAS for journeys that started with DSP but closed elsewhere. If DSP rarely closes but frequently starts high-value paths, you have evidence that pausing DSP would damage downstream Sponsored Products performance.

2. Identify wasted single-touch spend — In the Path to Conversion detail table, filter to paths where only Ct 1 is populated (single-touchpoint journeys) and check User Purchase Rate. Low purchase rates on single-touch paths from a given campaign type suggest that campaign needs reinforcement from other formats to convert.

3. Optimise frequency caps and channel mix — Use the Unique Users by Path Step chart to see how many shoppers are receiving 4 or 5 touchpoints. If a large share of high-frequency journeys is dominated by one campaign type, you may be over-serving with that format and under-investing in diversifying touchpoints.

4. Build a "closer" campaign list for budget pacing — Sort the First versus Last Touch view by Total Sales descending on the Last Touch column. The campaign types that close the most revenue are your protected-budget closers — never pause them before quarter-end.

5. Diagnose a sudden ROAS drop — When headline ROAS falls, open the Path to Conversion detail table for the affected period and compare against the prior period. A shift in which path sequences dominate (e.g. more single-touch Sponsored Products paths, fewer DSP-assisted ones) usually explains the change better than any single-campaign metric.

Limitations & Notes

• 14-day data lag. Data is fetched 14 days after a reporting period closes to allow Amazon's standard 14-day attribution window to complete. Results for a given period are available on day 15.

• Maximum five touchpoints. AMC reconstructs paths up to five steps. Shoppers exposed to more than five ads still terminate at step 5 in this model — the earlier touchpoints in very long journeys are not visible.

• Asymmetric ROAS denominator. Ad Spend in the table is summed across all touchpoints of the path, while ad-attributed Purchases and Sales are credited to the Last Campaign only. This is intentional — it shows true cost-to-convert — but it means path-level ROAS is not directly comparable to a single-campaign ROAS in other dashboards.

• AMC privacy suppression. Paths involving very small user counts are anonymised by Amazon as "redacted by Amazon" and hidden by default. Enable Show redacted to make totals reconcile, but individual redacted rows cannot be analysed further.

• Ad Product Type granularity only. Paths are reconstructed at the campaign type level (DSP, SP, SB, SD), not at the individual campaign level. Two different DSP campaigns appearing in sequence will show as DSP → DSP, not as two distinct campaigns.

• Last-touch convention for conversions. Even though the dashboard exposes first-touch data, the underlying Purchases / Sales values follow Amazon's standard last-touch attribution. The First versus Last Touch view does not re-attribute revenue to the first campaign; it groups last-touch revenue by what the first touch was.

• AMC integration required. This dashboard requires an active AMC connection. Contact your emax digital account manager if data is missing or the dashboard is empty.

Data Refresh

Data updates daily, with a 14-day lag after each reporting period ends. Results for a given period are available on day 15 after that period closes.