Activity at the start.
Campaign reporting showed spending and engagement. Those numbers helped describe the response to an ad, but we also needed to know which sources led people to act on an offer.
Client work · Social platform · 2026
Which campaigns led
to the next step?
For a social platform using Messenger to reach people, I designed the measurement that connected offer clicks back to the ad or post that started the conversation. That gave us a way to compare campaigns beyond clicks and conversation counts.
Delivered through a partnership with an advanced chatbot technology company. The end client is unnamed.
The measurement result
924 of 930 offer clicks
connected to their source.
From June 25 to July 21, 2026, the report assigned 924 of 930 recorded Facebook offer clicks to a specific ad or post using a first-touch attribution rule.
99.4%
Offer clicks with an
identified source
924 attributed6 without a matched source
Source: campaign attribution report, June 25–July 21, 2026. Offer clicks count link clicks, not unique people or purchases. The 99.4% figure measures attribution coverage under the report’s first-touch rule.
How the attribution was counted
People were assigned to their first recorded Facebook source. About 24% encountered the same creative as both a boosted post and an ad on the same day; the report used a first-touch tiebreak to assign those people.
The report states that test and bot traffic were excluded and that its click filter matched the executive reporting dashboard. The window covers the 27 days of retained logs available for this analysis.
The business problem
The ad report only told part of the story.
People could enter through an ad or a post, continue in Messenger and click an offer later. To decide where the budget should go, we needed to keep those steps connected.
A decision further along.
Connecting the offer click to its source made it possible to compare ads on the same downstream action. We could investigate where spending and results were out of balance.
What needed connecting
Keep the source with the conversation.
Ad or post
Where someone arrives.
Identify the campaign entry point, including the different paths into Messenger.
Messenger conversation
Carry that context forward.
Connect the conversation to its originating ad or post so the source remains available.
Offer click
Measure the next action.
Associate the recorded offer click with that source and compare the result with campaign spending.
What it helped us see
More people entering didn’t mean
more offer clicks.
The post-led path accounted for 3,800 first-touch people and 317 offer clicks. The ad-led path accounted for 1,635 people and 607 offer clicks. Measuring what happened inside the conversation gave us a different view from entry volume alone.
Click-to-Messenger ads
607 offer clicks
From 1,635 first-touch people
Post comments to Messenger
317 offer clicks
From 3,800 first-touch people
These are source groups under the report’s attribution rule, not a controlled test. People can click more than one offer. The comparison does not establish cost efficiency or return on ad spend.
The practical use
Compare the next action, then investigate the difference.
The report made it possible to see which ads and posts were associated with offer clicks. That gave us specific creative and entry paths to investigate and a basis for the next campaign test.
My role
Marketing decisions, backed by a system.
I managed paid acquisition, authored the attribution specification and built automated reporting. The attribution report connected recorded offer clicks with their assigned source.
Design the connection.
I wrote the specification for carrying campaign information into the conversation and connecting it to downstream actions. The work accounted for different entry paths instead of assuming every conversation arrived with the same information.
Manage the acquisition.
I owned paid acquisition, using the connected results to guide campaign analysis and budget recommendations. The measurement work served the decisions I had to make in the account.
Build recurring reporting.
I built an automated reporting agent to bring campaign performance into a recurring review. It helped surface changes and prepare recommendations for review as part of the ongoing marketing work.
Plan what comes next.
I also designed how the ad’s message could inform the offers shown in the conversation. That design work is separate from the reported attribution result described here.
Where this work helps
Can you connect your campaigns
to what people do next?
I work with teams whose advertising, customer experience and reporting need to connect. We start with the decision you need to make, then build the measurement to support it.
Talk about your measurement Related: connecting campaigns to enrollment quality →