Opportunities › High-performance OLAP database for analytical queries › Proposed app
Pacegrid
Replay your slow queries on your parquet files and compare engines side by side
- Score
- 40
- confidence 53
- Evidence
- 5
- items on the page
- Rivals
- 2
- named in sources
- First version
- 45 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2016
- Established market · AI estimate
ScreensConcept
Desktop app: Data engineers need to review dense benchmark results, compare multiple query engines, and analyze performance metrics across large datasets—work done at a desk with multiple screens. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Dataset
Connected102M rows
About this appAI proposal
Pacegrid is a query benchmark workbench for data teams. Point it at parquet files in your own storage, paste the queries your pipeline runs, and see how each engine performs on your data, not on a generic test set.
Public benchmark claims, such as a faster engine on a 14GB, 100M-row dataset, say little about your own schemas. Pacegrid turns that question into a repeatable run with latency, spread and a result check, so you can judge whether switching is worth it.
It is built for data engineers and developers who embed analytical databases in apps and pipelines. Every run produces a shareable report link for the people who approve the change.
- Tagline
- Test the speed claims on your own data.
- Voice
- PrecisePlainPractical
- Why this name
- Pace is how fast a query runs; grid is the table of results and the columnar data underneath. Two plain words joined.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Bring your own parquet
Connect read-only to a bucket or upload a sample. Pacegrid reads schema and row counts without copying the full dataset.
No member yet
Query replay
Paste queries or import a query log. Each one runs several times per engine so you see median and worst-case latency.
No member yet
Side-by-side engines
Run the same query on DuckDB, ClickHouse and any engine you add. Speed-up is shown per query and in total.
No member yet
Result match check
Row counts and checksums are compared across engines, so a faster answer only counts when it is the same answer.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#12332EDeep Pine · primary#F4F1E8Paper · surface#1B2321Ink · text#E8833AEmber · accent#2F8F83Teal Data · data
Typography
Aa
Pacegrid
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Test the speed claims on your own data.
PricingAI proposal
The material gives no competitor prices to base this on: ClickHouse lists prices on request. These tiers are a starting point to test.
Free
Free
- 1 dataset up to 5 GB
- 20 runs a month
- 2 engines
- Public report links
Team
$49 a month
- Datasets up to 500 GB
- Unlimited runs
- Private report links
- Regression watch
- 5 members
Scale
Contact us
- Private network access
- Single sign-on
- Custom engines
- Support
Would you pay Team · $49 a month?
Your answer tells PWAS which prices people accept. It changes no score.
No answers yet.
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- Shared outputs. Each run makes a read-only report link that engineers send to leads, vendors or open-source threads. Viewers see timings, settings and the Pacegrid name on the report.
- Do they become customers?
- Viewers are often data engineers who can run their own suite, but this is untested. Test how many report views turn into new sign-ups.
- First customers
- Data engineers at small software companies that embed an analytical database in their product and query 100M+ row parquet datasets.
- Where to reach them
- Hacker News: Post a free public benchmark of a popular dataset and invite people to run their own queries.
- Data engineering communities: Offer free runs in Slack and forum threads where people ask about slow queries.
- First milestone
- Fifty teams run a suite on their own data and five pay for Team within 60 days.
Why people already pay for this
- VerifiedHacker News post received 13 points and 10 comments
- Public claimExisting OLAP databases have performance limitations that create demand for faster alternatives
- Public claimBenchmarks on realistic analytical data sizes show meaningful performance gaps
- VerifiedClickHouse sells paid plans with prices on request, on its own pricing page, plus prices on request for larger customers
- Public claim
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · Query workersContainerised batch workers with pinned engine buildsRuns each engine on fixed hardware so timings are comparable and repeatable.
- This app · Object storage accessRead-only S3-compatible connectorsReads customer parquet files in place without storing full copies.
RisksAI inference
- CommercialTeams may run a benchmark once and never return, so a one-off tool may not support a subscription. Regression watch must prove repeat use.
- TechnicalFair comparisons need tuned settings and fixed hardware. Results that look unfair will lose trust quickly.
- DataCustomers may refuse to give cloud access to production parquet. Sampling and private-network options are needed.
- PlatformEngine licences and cloud compute costs per run could cut margins on large datasets.
- CommercialThe faster engine in the evidence is a public claim from one post. The product depends on faster engines existing and being worth testing.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.