Simulation
Content assembly
The same blocks, ordered differently for two labeled personas. Nothing here is a measured statistic.
Understanding the labHow to use it and why you'd want to
Why you'd want to
The same four blocks are shown in two preset orders, one for a marketer and one for an engineer. After Accept the page queues lab_interaction for the page view only.
How to use it
- Switch the persona and watch the block order change.
- The two orders are fixed. The blocks are the same.
- After Accept, only the page view is queued.
Simulation. Two labeled presets. The blocks are the same. The order is the rule. No model call and no stored profile.
Compare
Same blocks, two orders
Preset
First-time marketer
1. Intro
This site shows the signals it uses. The line does not name a person.
2. How
A rule reads a signal and picks a slot. The why lab names the rule.
Open3. Offer
The offer is the white paper. It is free and the same file for both personas.
Open4. Proof
Open the consent lab to see the gate. This block does not cite a number.
Open
Preset
Analytics engineer
1. Proof
Open the consent lab to see the gate. This block does not cite a number.
Open2. How
A rule reads a signal and picks a slot. The why lab names the rule.
Open3. Intro
This site shows the signals it uses. The line does not name a person.
4. Offer
The offer is the white paper. It is free and the same file for both personas.
Open
Why am I seeing this? The marketer preset leads with the intro and the how. The engineer preset leads with the proof link, which points at the consent lab and does not invent a statistic. Both end by offering the same white paper.
WalkthroughHow we built itShowHide
Example vendor setup is marked below. eVar and event numbers are examples. This lab does not call Adobe.
The content assembly lab shows the same four blocks in two preset orders, marketer and engineer. After Accept it queues lab_interaction for the page view, and the lab does not call Adobe.
Store two orders
assemblyOrder lists intro, how, offer, proof for the marketer, and proof, how, intro, offer for the engineer. The block ids are the same four.
Look up the order
blocksFor returns that array for the persona. It does not sort, score, or read the visitor.
Render the blocks in order
Each column maps the list to the same copy. Proof, how, and offer link to real pages. The offer is the white paper for both personas.
Queue the page view only
useLabView queues lab_interaction with action view and label page. This lab does not call trackLab for a block or a persona. The order is not a parameter.
From the code
Two fixed orders. blocksFor returns the list for that persona.
export const assemblyOrder: Record<AssemblyPersona, BlockId[]> = {
marketer: ["intro", "how", "offer", "proof"],
engineer: ["proof", "how", "intro", "offer"],
}
export function blocksFor(persona: AssemblyPersona) {
return assemblyOrder[persona]
}The column walks blocksFor and prints each block.
{blocksFor(persona).map((id, index) => {
const block = BLOCKS[id]
return (
<li key={id} className="min-w-0 rounded-md bg-muted p-3 text-sm leading-relaxed">
<p className="font-medium">{index + 1}. {text(locale, block.en, block.es)}</p>
<p className="mt-1 break-words text-muted-foreground">{text(locale, block.bodyEn, block.bodyEs)}</p>
{block.href ? <Link className="mt-2 inline-flex min-h-11 items-center underline-offset-4 hover:underline" href={localizePath(block.href, locale)}>{tr(locale, "Open", "Abrir")}</Link> : null}
</li>
)
})}
</ol>
</div>Google and Adobe
Google queues lab_interaction after Accept. Adobe is an example of the same fields.
Queued only after Accept. useLabView queues lab_interaction once, with lab_id content-assembly, action view, and label page. Choosing a column does not queue another event. The block order is not a parameter. This site does not call gtag('set') or Adobe.
Simplified shape of the object queued after Accept. Register any new parameter as an event-scoped custom dimension. This site does not call gtag('set').
dataLayer.push({
event: "lab_interaction",
lab_id: "content-assembly",
action: "view",
label: "page",
})Example implementation. This site does not call Adobe. eVar59, prop59, and event59 are example variable numbers. In Launch, a data element reads the value and a rule sends it when the value changes. In the report suite, enable eVar59 with Most Recent allocation and expire it on the hit. A processing rule is unnecessary if the data element maps straight to that eVar.
Example alloy payload. _example is a placeholder schema, not a real tenant.
alloy("sendEvent", {
xdm: {
eventType: "web.webinteraction.linkClicks",
_example: {
assemblyPersona: "marketer",
},
},
})Example hit. eVar59 is hit-scoped in this example. prop59 dies at the same hit.
s.eVar59 = "marketer"
s.prop59 = "marketer"
s.events = "event59"
s.linkTrackVars = "eVar59,prop59,events"
s.linkTrackEvents = "event59"
s.tl(true, "o", "lab_interaction")Translation
Each lab field, the GA4 name, and the example Adobe variable. Scope is the part that usually gets lost in the translation.
Page view
GA4. lab_interaction with action view and label page.
Adobe. Example eVar59, expire on the hit. The example value would be the action.
Scope. Event. It records that this page was viewed, not which column was read.
Persona
GA4. Not sent. marketer and engineer are the two preset keys.
Adobe. Example prop59 on the hit.
Scope. Hit, if you tracked a column. It is not a user property.
Block order
GA4. Not sent. The order is the array from assemblyOrder.
Adobe. Example eVar59 if you sent the first block id. This event does not.
Scope. A preset on the page. It is not a visit property.
Block id
GA4. Not sent. intro, proof, how, and offer are the four ids.
Adobe. Example prop59. Not sent.
Scope. Hit, if you tracked one block. It is not an event on this page.
Recommended next
Why these picksReading path transitions…