x-octo home Business judgment on AI products
中文

Real-revenue cases

Starting A Proofreading Business With A Global Client Base From 66

Real demand

Cambridge Proofreading runs an online proofreading and editing service staffed by 300+ freelance editors worldwide, serving students, academics, NGOs and businesses, with orders from 66 countries in 2022.

Founder interview, self-reported 素材未提及Google Ads 长尾关键词Facebook 再营销口碑转介自然搜索排名
Startup cost
$200
Primary source
View original

01

Real-demand verdict

Real-demand verdict

Real demand

It replaces asking a native-speaker friend or emailing agencies for custom quotes, turning a recurring need (non-native writers needing language polish) into a standardized, instantly orderable service. Twelve years of word-of-mouth repeat business shows people keep paying.

How it makes money

Clients pay per order for proofreading and editing, booked through a flat-rate transparent price list on the site instead of email quotes; editors are global contractors paid per job.

What old behavior it replaces

Asking a friend to fix your writing, or emailing agencies for a custom quote.

02

Where the first customers came from

The founder handed out flyers outside the student guild and got a local student inquiry the next day; early acquisition came from Google Ads and Facebook remarketing.

Acquisition channels Google Ads 长尾关键词Facebook 再营销口碑转介自然搜索排名

03

Tactics you can copy

  1. 01Publish flat-rate pricing on the site so clients order instantly instead of waiting for an email quote — an early edge over competitors.
  2. 02Strip the order form down to only the fields you actually need to cut friction.
  3. 03Put contact info, a registered address, third-party reviews and refund/confidentiality policies on every page to build trust without brand recognition.
Moving it to an AI business

For AI builders: the transparent pricing + minimal order form + trust-signal stack transfers directly to AI writing/editing tools or human-in-the-loop services, especially for new products without brand recognition. The case itself is not an AI business, so AI relevance is low.

04

Evidence and limits

Self-reported numbers are unaudited — treat them as leads, not facts
Founder interview, self-reported

Revenue basis · Founder self-reported, unaudited (see source page)

What evidence is missing

No revenue or profit figures at all — only scale indicators like 300 editors, 66 countries and 12 years in business.
Whether the 300 editors are contractors or full-time, and unit economics (average order value, editor cost share), are undisclosed.
No data on Google Ads or Facebook remarketing ROI, or on what share of clients come from referrals.