I thought I was building a scraper. Indie Hackers made me realize the product was the workflow.

Hey IH,

I posted here recently about a niche Apify actor I am building for commercial real estate data.

The actor pulls public listings from LoopNet + Crexi and returns one cleaner dataset with duplicate signals, source links, cap-rate context, days-on-market context, and broker fields when available.

My first instinct was to describe it as:

a LoopNet + Crexi scraper

or:

a cheap CoStar alternative for first-pass scans

After reading the comments, I think both framings need to be handled more carefully.

The first mistake: calling it a scraper
Technically, scraping is part of how the product works.

But that is not what the buyer cares about.

A broker or analyst does not wake up wanting:

scraping
pagination
deduplication logic
normalization
exports
They want the result:

one clean, structured market file
Something they can actually open, filter, sort, export, and use.

That was the clearest positioning lesson from the feedback:

The product is not the scraper. The product is the clean file the workflow produces.

That sounds obvious, but it changes the way I talk about everything.

The second mistake: being too loose with "CoStar alternative"
This is the other point I want to be honest about.

CoStar is a huge, mature, enterprise-grade commercial real estate platform.

It has proprietary data, deep history, research workflows, sales teams, integrations, and a very different level of coverage.

My product does not replace that.

It would be ridiculous to pretend that a small Apify actor is "competing with CoStar" in the broad sense.

The more honest positioning is narrower:

Not a CoStar replacement. A cheaper way to turn public LoopNet + Crexi listings into structured data.

That is the wedge.

Not "we do everything the incumbent does."

More like:

For this one public-listing workflow,
can we make the data cleaner,
faster to use,
and much more affordable?
That feels both more credible and easier to sell.

What the product actually does
The workflow is simple:

Input:
market + listing filters

Sources:
LoopNet + Crexi

Output:
structured market file
The output is the important part.

It is meant to be easier to work with than raw portal browsing:

one row per listing
source links preserved
duplicate / cross-posting signals
normalized cap-rate context
days-on-market context
broker fields when visible
CSV / Excel / JSON / API export
The value is not that the actor "scrapes."

The value is that the user gets a cleaner file without rebuilding everything manually in a spreadsheet.

The pricing matters too
Another thing I probably under-explained:

The low-cost angle is not just a marketing trick.

It is part of the product.

For a lot of first-pass market research, people do not necessarily need a heavy enterprise platform.

They need a quick, affordable way to answer:

What is listed?
What is duplicated?
What looks stale?
Who is attached to the listing?
What source did this come from?
Can I export this cleanly?
That is why the pricing is around:

~$5 / 1,000 listings
The point is not to be the most complete CRE intelligence platform.

The point is to make one annoying public-data workflow cheaper and more usable.

The builder lesson
The main lesson for me:

If you are building a vertical data product, do not position the mechanism as the product.

In my case:

Mechanism: scraping LoopNet + Crexi
Product: clean structured CRE market file
Promise: affordable public-listing intelligence
That is much clearer than trying to sound bigger than the product really is.

I think developers and founders can smell when positioning is inflated.

So the better move is probably to be specific:

here is the exact workflow
here is what it does
here is what it does not do
here is why the price makes sense
here is a sample output you can inspect
That is what I am trying to improve now.

The actor is here for context:

https://apify.com/kazkn/commercial-real-estate-brokerage-intel?fpr=8fp2od

My question for IH:

For a niche B2B data product, what would make the positioning feel most trustworthy?

a real sample dataset by market
a side-by-side "messy portals -> clean file" demo
a transparent output schema
clear limitations and data-quality notes
pricing proof / cost comparison for the narrow workflow
Curious how you would communicate this without overclaiming.