The hidden costs of asking for quotations in the data market
Working towards faster, cheaper and safer data projects
About Data Boutique
Data Boutique is a data marketplace focused on web scraping. We make it simpler to match those who collect data with those who know how to use it.
The hidden costs of asking for quotations in the data market
Asking a vendor for a quotation is a back-and-forth hidden-costs-generating process with many inefficiencies.
Building complex solutions involves asking vendors for packages that are not priced yet
Asking for a quotation in B2B projects is often unavoidable.
Whenever building new or tailor-made solutions, we need to buy stuff in packages that have not yet been priced, produced, or assembled the way we need them, and data is a frequent case.
Often, quotations refer to components of larger projects, which also need to provide an estimate.
Examples: A firm bidding for a tender and asking a vendor for data collection costs; the IT department researching the market after being asked by Marketing for a BI solution; a startup evaluating the economics of different technology stacks to adopt.
It all revolves around trying to understand the cost of something that has not (yet) a price tag on it.
Quotations are a cost for sellers…
The process represents a cost for the sellers: The time spent on poorly written briefs and designing packages to address a request they have little information on ultimately builds up their operating costs.
Once the quotation is submitted, the waiting for the feedback of the entire decision chain begins, made of reminders, check-in emails, and rescheduled calls.
Then come scope revisions, alternative scenarios requests, and bulk discounts. Despite this, as often happens, the project might not even be won.
This is part of the CAC (Customer Acquisition Costs), eventually paid by buyers, as it will be factored into the price.
… and for buyers too
Inefficiencies also hit buyers: Time and resources spent writing down the requests, reading and scoring responses, and asking for revisions and alternative scenarios.
A long quotation-asking process means it takes longer to buy stuff, which means either we buy less frequently (and build fewer solutions) or hire more staff to do it. Both cases kill ROI.
From whatever perspective you look at this, it’s a dollar-killing inefficiency.
Unbearable for small-size trades
While this is acceptable in illiquid markets (as illiquid as a once-every-five-year-million-dollar deal), it becomes an unbearable friction for small to mid-size trades (hundreds or a few thousand dollars multiple times a year).
Web scraping data - the business we’re in - falls in the latter case.
How we solved it: Schemas and Bundles
Since our mission is to make access to data faster, cheaper, and safer, solving this money-burning problem was a priority.
What we did was the following:
Standardizing the building blocks
Empowering buyers to play before they buy
1. The building blocks (Schemas)
Simplifying starts from the foundations: To make the process smoother, we must identify simple, modular building blocks for our data use cases. The tricky part is to simultaneously be simple and valuable.
Standard schemas provided the answer: Buyer-oriented data structures that serve “atomic” data needs. More complex solutions can be delivered just by assembling more schemas together.
Working in web scraped data made this part easier: Websites usually come in standardizable categories, such as e-commerce, classifieds, store locators, travel and booking, and so on.
When you buy an “E0001 Schema” you know what you get, regardless of the website. This took so much complexity from the equation and made life easier for everyone: Buyers and sellers.
2. Play before you buy (Private Bundles)
Once we had modularity, creating complex configurations was its natural evolution:
Data requests like “a dataset to monitor daily prices and promotions of H&M products in France” or “a dataset to measure inflation in Germany, the UK, the USA, Canada, China, and Japan for all Kering brands” is an exercise of assembly.
Inspired by tools like the AWS Pricing Calculator, we created Private Bundles, a feature where users can simulate their data packages.
With Private Bundles, users can change the scope and frequency, assemble atomic blocks as they need, save their selections, and share them with their team, with no commitment to buy.
Private Bundles are a safe environment where users can experiment different ideas before committing to buy
A Private Bundle enables to:
Start planning on initial ideas and saving them for later, defining data specs as you go and keeping draft versions as more information comes in;
Have a real-time cost projection, and see how this changes when you choose different data types, vendors, data refresh frequency, and content;
Factor in volume discounts, as they are automatically inserted in the calculation;
Contribute with other team members to its definition, sharing the final draft with the decision-makers.
Improve the speed of thinking data projects
The speed of project design improves significantly as a buyer can formulate, correct, and rephrase their project brief directly on the platform, obtaining a real-time cost quotation.
To save even more time, you can start from a Public Bundle, clone it, and work from there: Example of a public bundle.
Often, buyers need to evaluate multiple scenarios, inclusive of data costs, before they can make a decision.
We make this process faster.
About the Project
Data Boutique aims to increase web data adoption by creating a win-win environment for data sellers and buyers. Join our community, it’s free. More can be found on our Discord channels.
Thanks for reading and helping our community grow.