"Your PDP dropped from position 3 to 9 on Éxito. It wasn't price. It was that the competitor added 4 images and 12 attributes last week." If you could have this level of clarity on every ranking move across your portfolio in LATAM, how many different decisions would you make every week? That's the question this series tries to answer.

The way CPG brands manage their PDPs in LATAM looks like driving a car with the mirrors covered. You know where you want to go, you sense something moving around you, but you can't see precisely what's happening to you or why. The retailer shows you a dashboard when it suits them. The agency shows you another. Each one tells a story where they come out looking good.

The problem is that neither of them can honestly tell you why your product ranks the way it does — because neither is independent. If organic position is the most reliable metric for whether your ecommerce investment is working (and we'll argue it is), then you need someone with no interest in inflating it to measure it for you. That's the sentence that anchors the rest of the article.

This guide opens three ideas we then detail in the two following blog articles: (1) your PDP is being judged by two engines simultaneously — the retailer's and Google's — and each one rewards different things; (2) the agentic shift isn't a future scenario, it's already happening, and it multiplies the surfaces you need to monitor; (3) the only useful measurement of your PDP is the one no retailer or agency controls.

The PDP that's falling and why you can't diagnose it

Imagine a real case. Your top SKU in a competitive category — a detergent, a yogurt, a shampoo — is ranking stable at position 3 on Éxito for months. One Monday morning the report from your platform tells you it dropped to position 9. Your first reaction is to check price: did anyone lower? No, prices are the same. Did something happen with stock? No, all in order. Did the retailer's algorithm change? Impossible to know — they don't tell you.

Four weeks later you're still at position 9 and you couldn't diagnose the cause. The agency suggests raising the retail media bid. The retailer offers to sell you a featured placement. The ecommerce team is putting together a "content optimization" plan without knowing what to optimize.

The likely reality, based on an audit we did on a major brand, was this: the competitor now ranking above you added 4 new images and completed 12 attributes from the category schema the week before the drop. Your PDP didn't change. Theirs did. The retailer engine detected superior completeness and reallocated visibility.

You can't diagnose that with the retailer dashboard (which only shows you) or the agency's (which sells media, not intelligence). You need a view that watches every PDP in the category — yours, your competitors' — every day, and identifies exactly what changed and when. That's the nature of an independent measurement.

The two engines reading your PDP

Your PDP at Éxito, Jumbo, Olímpica, or any LATAM retailer is being evaluated by two engines at the same time. Optimizing for one and ignoring the other leaves positions (and sales) on the table.

Engine A: the retailer's search. Each retailer has its search algorithm that orders organic results when someone types a keyword. Éxito, Carulla, Jumbo Colombia, and Olímpica run on VTEX, with configurable engines that reward attribute completeness, title structure, image count, review volume, and conversion rate. Each with specific weights: Éxito and Carulla removed filter facets recently, so discovery happens via keyword search + categories and attributes still feed the ranker but are no longer a shopper-facing filtering lever; Olímpica keeps facets active and, combined with typically less-populated PDPs, is the zone with highest ROI per hour invested in attributes; Jumbo Colombia also keeps facets live, rewards longer titles, and slots 1-3 are dominated by sponsored, so organic content fights harder for slots 4-10.

Engine B: Google. The same SKU is indexed by Google through three parallel surfaces, each with its own rules:

  1. Traditional organic search — rewards schema markup, page experience, and backlink authority.
  2. Google Shopping and free listings — depends on the Merchant Center feed and consistency between feed and the live PDP. Price mismatch between the two = suppression.
  3. AI Mode and AI Overviews — the new front. Rewards structured completeness over keyword stuffing. The metric emerging in the industry is "Golden Record" — listings near 100% attribute completeness are seeing several times more visibility in conversational answers. Aritzia reported an 80% lift in revenue from optimizing for Google AI Max (coverage: Search Engine Land).

Your ecommerce team is probably focused on Engine A. Your digital marketing team is probably focused on part of Engine B. Few brands in LATAM are systematically measuring both — and almost none are measuring AI Mode.

The independence question

There's a structural point that's discussed too little but that defines the entire PDP measurement problem. The two parties with the most data on your PDP performance are the retailers and the agencies — and both have systemic conflicts of interest.

The retailer is at once the operator of the engine that ranks your product AND the media network that sells you advertising. When it shows you that your product ranks at such-and-such position, it's showing a metric calculated by the same platform that charges you to appear. Initiatives like Kroger's PrecisionView 360 in the US are attempts to improve this, but they're still measurement produced by an interested party.

The agency has a similar but different incentive: it needs to demonstrate that its management is paying off. If it tells you "look, organic position rose because we did X," that X always coincides with what they sold. As an agency-side commenter on a recent forum put it: "the biggest pain right now is proving incrementality. And of course retailers own that data, so there's basically no third-party verification".

What's missing? An independent third party whose only job is to tell you the truth — including when that truth is "your position isn't improving," or "what improved wasn't because of the agency, it was because the competitor went out of stock." That's the position ePerfectStore occupies in LATAM: daily PDP scraping at Éxito, Jumbo, Olímpica, Rappi, with no media sales, no retailer contract, no agency alignment.

The 3 layers of PDP diagnosis

A useful PDP measurement needs to answer three different questions:

Layer 1 — the "what" (diagnosis). A PDP Health Score per SKU per retailer, with specific sub-scores for each component each algorithm rewards: title structure, long description, image set, video, structured attributes, reviews, Q&A. Each one scored against the benchmark of the top 5 products in each category.

Layer 2 — the "vs. who" (competitive gap). For each tracked keyword, side-by-side comparison with the top 5 organically ranked PDPs. "Your title has 47 characters; the top 3 average 78. You're missing the words 'salt-free,' 'oily hair,' and 'women' — all in the titles ranking above you."

Layer 3 — the "what to do" (prioritized actions). Recommendations ordered by estimated rank-lift, not alphabetically: "Add 3 keywords to the title → estimated +4 positions on Éxito. Upload 4 missing images (currently 2/6) → estimated +2 positions."

Without all three layers, you aren't diagnosing — you're monitoring with more detail. And monitoring isn't the same as actionable intelligence.

Alerts that click: the "It wasn't X. It was Y." pattern

The alerts worth having follow a consistent pattern that feels more like intelligence than dashboard. Three steps: observed movement → obvious cause ruled out → specific, fixable gap. Five examples, each the kind of alert the system pushes when something crosses a threshold:

"Your PDP dropped from position 3 to 9 on Éxito. It wasn't price. It was that the competitor added 4 images and 12 attributes last week."
"You lost 6 positions on 'lactose-free milk' on Jumbo. It wasn't your content. It was that the competitor added 47 reviews in 10 days — probably with a sampling campaign. Your PDP has 12 reviews from 8 months ago."
"You've been at position 8 on Olímpica for 'liquid detergent' for three weeks. The 7 PDPs above you all have a video. Yours doesn't. It's the only variable you're losing on — same price, better rating, more images."
"Your stage-3 diapers PDP dropped from page 1 to page 2 on Olímpica. You didn't change anything. The retailer activated a new 'fit type' filter last week — competitors filled that attribute. You have it empty and no longer appear when buyers filter."
"Your top SKU ranks #2 on Jumbo but #14 on Éxito with the same content. The difference: your title has 45 characters. Éxito rewards titles 70+. It's a 30-second change that can move you 8 positions."

Each one delivers three things: a movement the team is already feeling, a specific cause that rules out noise, and a concrete action. That's what differentiates actionable intelligence from a panel with lots of numbers.

The agentic shelf isn't the future: it's happening now

The last relevant change for this series is the agentic shift. ChatGPT, Perplexity, Gemini, and the new assistants from Mercado Libre, Amazon (Rufus), and local retailers are mediating more and more purchase decisions. The industry talks about the "infrastructure era of retail media" — and the strongest claim coming out of that conversation is that agentic search no longer rewards keyword matching but intent matching.

This changes what "optimizing the PDP" means. It isn't just completing attributes for the retailer's filter to include you. It's writing content that a model can map to the shopper's intent. A diaper PDP that says "size 3, 40 units, hypoallergenic" is complete. A PDP that says "for babies 6-10 kg who are actively moving; elastic fit for crawlers; pediatrician-recommended for sensitive skin" is intent-legible. Same product. The second maps naturally to "what diapers for my baby who's already crawling." The first doesn't.

That distinction — intent legibility — is the new metric brands need to start measuring, and we cover it in detail in the third article in the series on Intent Coverage and the Google engine.

Authenticity: the bridge between the retailer engine and the agentic engine

There's a question that sits underneath both layers and that almost nobody discusses out loud: does the content you're optimizing for the ranker feel real to the shopper who lands on it? Tom Cohley, CEO of Cohley, frames it this way in a Digital Shelf Institute analysis (September 2024): "You can't fake authenticity." And he adds: "You think about every touch point that goes on there, whether it be online or offline, and you need to be thinking about the content that you're putting forth to usher them through that journey and give them the confidence, trust, and education they need to feel confident in making a purchase."

The data point that ties this to hard metrics comes from the same article: per BazaarVoice data, products with fewer than 50 reviews suffer a minimum 30% drop in conversion rate vs. peers with sufficient volume. And the Nature's Way case shows a 20% conversion lift on Amazon after implementing an authentic UGC strategy.

Three operational rules that follow:

It's the 2026 version of the old retail principle: shoppers buy when they trust. What changed isn't the rule — the witnesses changed.

In summary

Question Short answer
Who judges your PDP? Two engines simultaneously: the retailer (Éxito/Jumbo/Olímpica) and Google (organic + Shopping + AI Mode).
Who measures that today? The retailer and the agency — both with systemic conflict of interest.
What's missing? An independent, daily, cross-retailer, cross-surface measurement.
What does good intelligence look like? Pattern: "movement → rule out obvious cause → specific, fixable gap."
And the agentic shift? Already happening. New metric: Intent Coverage, not keyword matching.

To recap: your PDP is being judged by two engines simultaneously, those engines reward different things, and the only useful measurement of how they're reading you is the one an independent third party can deliver. In the next article we go deeper into the retailer engine — what each LATAM retailer rewards, where the most common gaps are, and what a well-done diagnosis looks like. In the third article we cover the Google engine and the new Intent Coverage metric for conversational search.

Read next: The retailer engine: how Éxito, Jumbo, and Olímpica decide your organic position · The Google engine and the new metric: Intent Coverage for conversational search.

Sources

Want to know why your PDP ranks the way it does — without going through the retailer dashboard or the agency? ePerfectStore.com measures both engines independently, every day, at every LATAM retailer.

Back to Blog