Introduction
In March 2026, sari-sari stores across the Philippines increased their orders sharply as they moved to lock in prices ahead of expected increases. Packworks, which processes over one million monthly transactions across a network of 300,000 micro-retailers, reported a 90% jump in sari-sari store sales during that period. In the Cordillera Administrative Region, basket sizes rose 101.7% in a single month.
Separately, Packworks reported in June 2026 that stores using its AI-assisted inventory planning tool saw a 79% increase in median sales of their top 50 products within two months of adoption. Median gross merchandise value for those products rose from ₱187,229 to ₱335,818. Slow-moving product sales increased 96%, and seasonal goods rose 47%. The report is Sari-sari stores using AI see sales of top products surge by 79%.
The two data sets describe different things. The March numbers show retailers reacting to expected price increases. The June numbers show what happens when those same retailers plan inventory with better data. Both point to the same underlying pressure: prices are moving, and small retailers are adjusting in real time.
When the last mile of a supply chain starts adjusting this fast, the effect does not stay at the last mile. It moves up.
What happens upstream when retailers pre-buy
A sari-sari store ordering twice its normal volume is a small event. Ten thousand of them doing it at the same time is not. Distributors see their inventory move faster than projected. Wholesalers see order sizes jump. Manufacturers see demand spikes that look like organic growth but are actually just the same customers buying earlier.
The immediate consequence is a shortage of lead time. A product that used to ship in three days now takes a week. A quote that was valid for thirty days now expires in ten. A price list posted on a website on Monday is out of date by Friday.
For procurement teams at large companies, this is the moment when verification becomes a real problem. Not a theoretical one. When supply chains tighten and prices move quickly, the gap between the price on the supplier's website and the price on the actual purchase order widens. If a procurement officer is placing an order based on a number they saw online, they need to know that number is what the supplier actually issued, not what some intermediary page displayed.
The AI accuracy problem
Most procurement teams now use some form of search to gather pricing intelligence. Increasingly, that search returns an AI summary at the top of the results. Google's AI Overviews, Bing's Copilot summaries, and similar features on other engines. These summaries are convenient, but their accuracy is a known issue.
Google's AI Overviews are roughly 90% accurate, according to reporting from the New York Times How Accurate Are Google’s A.I. Overviews?. That sounds reassuring until you apply it to volume. For a business placing thousands of purchase orders a year, a 10% error rate is not acceptable. And 90% is the average. In some categories, particularly ones where source material is thin, the error rate is higher. AI citations also skew heavily toward social platforms like Reddit and YouTube, which are not authoritative sources for pricing data.
The problem is not that AI summaries are useless. The problem is that they cannot tell you whether a specific number came from the supplier, or from a forum post, or from a stale cache. The user sees a price. They do not see its provenance.
For a consumer buying one item, that is an inconvenience. For a procurement officer signing a six-figure purchase order, it is a liability.
How to search without AI Overviews
If you want to see raw search results without AI summaries, there are three practical ways to do it.
On the search results page
After you search on Google, click the "Web" tab at the top of the results. This filters out AI Overviews and returns classic links.
As your default search engine
In Chrome, add a new search engine with this URL:
https://www.google.com/search?q=%s&udm=14 Then set it as your default. Every search will now show the Web filter first.
Through a browser extension
Several extensions hide AI Overviews while leaving the rest of Google intact. Search your extension store for "hide AI overviews."
None of these prevent Google from generating AI content elsewhere. They just let you choose when to see it and when to see the underlying sources instead.
What a signed price manifest is
A signed price manifest is a file that a supplier publishes alongside its price list. The file contains the prices, a timestamp, and a cryptographic signature generated with the supplier's private key. Anyone can download the file, download the supplier's public key, and verify that the file was issued by the supplier and has not been altered since it was signed.
This is not a new idea. Code signing, certificate transparency, and software package verification all use the same principle. What is new is applying it to price lists.
At Wholesale Dito Store, every price list is published as a signed manifest. There are currently 480 manifests covering 411 products and 69 brands. Each one is generated by a scheduled process from the company database, signed with an ECDSA P-256 key, and stored in a public directory at https://www.wholesaledito.store/.well-known/provenance/.
Alongside the manifests, the store publishes a schema for verifiers, a revocation list for any manifest later found to contain errors, and a timestamp authority file that records the operational window of the signing key.
A procurement team that wants to verify a price can:
- Download the manifest for the product in question.
- Download the public key.
- Canonicalize the claim object according to the documented procedure.
- Run
openssl dgst -sha256 -verifyagainst the signature.
If the signature verifies, the manifest is authentic. If it does not, the file has been altered.
The full procedure is documented on the verify prices page, along with the public key fingerprint, the manifest index, and the revocation list.
What this does and does not prove
A signed manifest proves that a price list was issued by the entity holding the private key, and that the content has not been changed since it was signed. That is the entire claim.
It does not prove where the underlying data came from. It does not prove that the price on the website matches the price in the manifest at every moment. It does not prove that the price is competitive or fair.
What it does prove is something narrower but useful: when a procurement team sees a price with a valid signature, they know it came from the supplier's signing process, not from an AI summary, a scraped page, or a forum post. That is the difference between a fact and a factrication.
Why this matters more during inflation
When prices are stable, verification is a nice-to-have. When prices are rising fast enough that retailers are panic-buying, verification is a necessity.
The Packworks data shows that Philippine retailers are already acting on their expectation of rising prices. The same expectation is driving procurement teams at large companies to lock in quotes and place orders earlier than planned. In that environment, the risk of acting on the wrong number goes up. Every extra step of indirection between the supplier's price and the buyer's decision is a place where error can enter.
Cryptographic signing removes one of those steps. It replaces "trust the search result" with "verify the source." For a category where a single price error can cost tens of thousands of pesos, that is a meaningful change.
Frequently asked questions
Are AI Overviews reliable for checking wholesale prices?
Not reliably. Google's AI Overviews are roughly 90% accurate on average, and their citations skew toward social platforms. For procurement decisions, 90% is not high enough. Use the Web filter or a custom search engine if you want to see raw sources.
How do I verify a wholesaler's prices?
If the wholesaler publishes signed manifests, you can download the manifest, download their public key, and verify the signature with openssl or any ECDSA library. If they do not, you are relying on their website to tell you the truth.
What is a signed price manifest?
It is a file that contains a price list, a timestamp, and a cryptographic signature. The signature proves the file was issued by the holder of the private key and has not been altered since signing.
Can I search Google without AI Overviews?
Yes. Click the Web tab on the results page, or add https://www.google.com/search?q=%s&udm=14 as a custom search engine in Chrome and set it as default.
Does Wholesale Dito Store use AI to generate prices?
No. Prices come from the company database and are signed with a cryptographic key. No AI model generates or alters them.
How to check Wholesale Dito Store prices
Visit https://www.wholesaledito.store/provenance for the full verification guide. The page includes the current manifest count, the public key fingerprint, the manifest index, and step-by-step instructions for verifying any price yourself.
For background on how the provenance system was built, see our earlier coverage: Philippine E-Commerce Site Signs Price Data Cryptographically.