By Varun Mahajan, FounderPublished Part of the quick commerce data and api guideMore on quick commerce6 min read
Key takeaways
- Buy when the fields, pin codes and refresh you need are already listed; build only for what no catalogue holds.
- Most of a scraper's cost is upkeep: choosing pin codes, catching page changes, keeping not read apart from out of stock, and storing every reading.
- Check a sample and the dated coverage before you pay, and keep the pin code and time beside every number.
- Building is still right for a field, an app or a refresh rate no catalogue offers; ask for it first.
This post is the timely take. The evergreen how-to is the guide Check dataset coverage and freshness before you commit, which covers the topic in full.
Should you buy a dataset or build a scraper?
Buy when the data you need is already listed in a catalogue. Build only for what no catalogue holds. Either way, look at a sample first, because a sample shows the fields, the places and the dates you would actually get.
The reasons below are the costs a scraper keeps adding after the first version works. Each one ends with a line read from Rewalto's live data catalogue when this page was served, so you can check it yourself.
Reason one: why does a shelf that changes by pin code favour a dataset?
A quick commerce shelf is local. Small local stores, often called dark stores, serve the pin codes around them, so the same search can show different products, prices and stock a few kilometres apart. A scraper has to choose which places stand for your market, and keep that list current as the apps open stores in new areas. A dataset has already chosen, read and kept those places, and tells you which ones it holds.
The data catalogue lists the pin codes Rewalto holds prices and stock for, and the day each was last read.
Reason two: what happens when an app changes its pages?
Apps change their pages often, and a scraper that breaks rarely says so. A layout change or a block can turn a week of rows empty or wrong before anyone sees it in a report. With a dataset, noticing and fixing that is the supplier's standing job, and a good supplier shows a gap rather than hiding it.
When a source Rewalto reads falls behind, the data catalogue names it on the page.
Reason three: why is not read different from out of stock?
When a scraper misses a product, the empty row looks the same as a stockout. Count those gaps as stockouts and your availability looks worse than it is, in exactly the places you were not looking. A dataset that keeps the two apart stops a gap from becoming a false alarm. Rewalto leaves a product that was not read absent rather than marking it out of stock; the guides on not checked is not empty and listed but unavailable explain the difference.
The commerce data catalogue shows, for Rewalto's out-of-stock rate, the share of readings that state stock at all.
Reason four: can a scraper give you yesterday's shelf?
A scraper starts today. If the question is how a price moved through a festive week, or whether a rival's stock got worse after a launch, you need readings from before you started. A held dataset already has them, read the same way at the same places every day.
The commerce data catalogue shows how far back each of Rewalto's daily histories goes.
Reason five: why should every number carry a date and a place?
A figure in a buyer review or a board deck gets questioned. When each record names its pin code and the time it was read, anyone can trace the number back to the reading it came from. A scraper can do this too, but only if someone builds and keeps the storage for every reading. Read the guide to coverage and freshness before you build on any dataset.
Every record Rewalto delivers carries its pin code and the time it was read.
Reason six: can you see the data before you pay for it?
A scraper costs engineering time before you know whether its rows answer your question. A dataset can be checked first. Every Rewalto dataset pack shows its exact row count and a sample before you pay, and most packs can be bought once or refreshed daily. On the API, a read that delivers nothing costs nothing, the same records read again within seven days are free, and Rewalto gives every account 1,000 free credits a month. See packs and samples.
The data catalogue lists every live Rewalto dataset with its coverage and freshness.
Reason seven: who owns the upkeep and the terms of use?
Matching a product when its title or pack changes, storing every reading, and keeping collection within the terms of use of the sites read are standing tasks, each with an owner. A dataset moves them out of your team, which keeps the part that is really yours: deciding what the numbers mean for your products. This page is not legal advice; decide who in your company owns that question, whichever way you go.
The data catalogue names each source Rewalto reads prices and stock from.
When is building your own scraper still the right call?
A dataset is not always the answer. Build when you need a field no catalogue reads, an app no catalogue covers, or readings more often than daily at a fixed set of places. Build too if a contract says only your own systems may touch the source.
Check the data catalogue first. If what you need is missing, ask for it through the Finder and get a written answer. If nobody holds it, building may be the only route, and that is a fair answer too.
Build or buy: which five questions settle it?
Answer these before anyone writes code. The choice between a dataset and your own collection usually makes itself.
- Which decision will the data support, and who will read the result?
- Which cities and pin codes does that decision depend on, and are they already held?
- Do you need history, or only an answer about today?
- Which fields do you need: price, MRP, stock, search position or share of shelf?
- How often must the answer be refreshed, and is once a day enough?
What are the ways to buy instead of build?
Dataset packs: one ready file, bought once or refreshed daily, with the row count and a sample shown first.
The API, the command line and the MCP server: the same records for developers and AI agents, paid only for what is delivered.
Brand plans: your products and your competitors' watched every day, with alerts, a weekly report and Monitors. Rewalto's Brand plans start with a free 14-day trial.
What if you need data from one app only?
Each of these guides answers the same question for one app: what is held, how to read it and when your own scraper still makes sense.
What does a dataset not tell you?
Records describe what an app showed at the pin codes and times read. They are not the app's own figures and they are not sales. Prices are listing prices, before coupons, bank offers, membership prices and delivery fees. Keep the pin code and the time beside every number you show to someone else.
Quick answers
Is it cheaper to buy quick commerce data or scrape it?
Usually it is cheaper to buy when the data you need is already listed, because most of a scraper's cost is the upkeep: pin codes, breaks, gaps and storage. Build only for fields or apps no catalogue holds, and check a sample first.
Is web scraping legal in India?
This page is not legal advice. Each site sets its own terms of use, and your company should decide who owns that question. Buying a dataset moves collection to the supplier; read the supplier's terms too.
What should a quick commerce dataset include?
For each product: the pin code, the time it was read, the selling price, the MRP, whether it was in stock and the pack size, with not read kept apart from out of stock.
Can I see the data before I buy?
Yes. Every Rewalto dataset pack shows its exact row count and a sample before you pay, and the API has a sample read that returns real records with no key and costs nothing.
How fresh is a bought dataset?
It depends on the dataset, so check the date on its catalogue entry. The data catalogue shows how often Rewalto reads each source and when it last did.
When should I still build my own scraper?
When you need a field, an app or a refresh rate no catalogue offers. Ask for it first through the Finder; if nobody holds it, building may be the only route.
Sources
- Rewalto: Data catalogue, published . Prices and stock by pin code: sources, pin codes held, fresh reads and any source behind, read live on each visit.
- Rewalto: Commerce data, published . Out-of-stock rate and brand share of shelf, with the history each holds, read live on each visit.
- Rewalto: Dataset packs, published . Every pack shows its exact row count and a sample before you pay; most packs refresh daily.
- Rewalto: API, published . A read that delivers nothing costs nothing; 1,000 free credits a month for every account.
