When I started reviewing AI companion apps, the process was pretty straightforward.
I'd pay for a subscription, use the app for a while, take notes, and write about what I experienced.
Most of the companies had no idea I was reviewing them.
That has slowly started to change.
As the site grew, I started reaching out to more companies before or during testing. Some were interested in affiliate partnerships. Some agreed to link back to our site after the review. Some offered free subscriptions or credits so I could test their product.
And recently, I noticed something about my own work.
The companies I was already talking to were getting better reviews from me.
Not more positive.
More detailed.
When I'm reviewing an app with no contact with the company, I'm basically doing what any customer would do.
If something doesn't work, I note it. If something is confusing, I write about it. If I like a feature, I explain why.
But when I'm already talking to someone from the company, I find myself going further.
If I encounter a bug, I'm more likely to try reproducing it.
If something confuses me, I spend more time figuring out exactly why.
If I think a feature could be improved, I don't just say I didn't like it. I start thinking about what I would change.
Strangely, I probably document more problems with products from companies I'm talking to, not fewer.
There's someone on the other side who might actually do something with that information.
Eventually, I started sending some companies a separate PDF after testing.
Bugs I encountered. Things that confused me. Features I liked. Things I thought could be improved.
At some point I realized I'd accidentally started doing a small usability test.
I'm signing up without help. I'm paying for the product. I'm figuring out the interface myself. I'm spending real time using it. And I'm documenting where the experience breaks down.
For a large company, feedback from one small review site probably isn't going to get much attention.
Smaller startups have been different.
They reply. They ask questions. Sometimes they're genuinely interested in what went wrong.
And that made me rethink what the relationship between a review site and the companies it covers has to look like.
I don't want to pretend this is purely altruistic.
We're building a site and we want it to grow.
We use affiliate links. We also ask some companies whether they'd be willing to link back to our site after the review.
And yes, that affects what we prioritize.
If I'm choosing between two apps and one company is actively working with us and willing to link back while the other isn't responding, I'm probably going to work on the first one sooner.
That doesn't mean the second app won't get reviewed.
And a backlink doesn't determine what goes into the review. If I encounter a bug, it stays. If I think something is bad, I say so.
But I've realized the relationship does affect something else:
how far I go beyond the review itself.
When I know my feedback might actually reach the people building the product, I'm more likely to investigate a problem further, document it properly, and think about how it could be improved.
The review still needs to tell the reader what I experienced. But now I'm also thinking about whether what I found could be useful to the people building the product.
A few companies have offered to cover my subscription or give me credits for testing.
It's tempting.
Testing enough products gets expensive, and there's nothing inherently wrong with a company giving a reviewer access.
But so far, I've preferred paying.
The reason is simple: I'm trying to review the experience of being an actual customer.
Paying is part of that.
I want to see the pricing before I buy. I want to go through checkout. I want to see what actually unlocks. I want to decide whether what I received was worth what I paid.
There are things you can miss when you're given special access.
Paying doesn't magically make me unbiased. There are still affiliate commissions, backlinks, relationships with founders, and other incentives surrounding the work.
It just preserves one part of the process I care about:
I enter through the same front door as the person reading the review.
I used to assume that being independent meant keeping the companies I reviewed at a distance.
I'm less convinced of that now.
Working with a company can introduce incentives. I think it's important to acknowledge that.
But I'm also finding that the relationship can make the work more useful.
The company gets detailed feedback from someone who actually used and paid for its product.
We might get an affiliate relationship, a backlink, or simply better communication.
The reader still gets the bugs, complaints, and things I didn't like.
The distinction I'm trying to maintain is this:
The review itself isn't what we're trading.
A backlink shouldn't buy a positive review. A higher affiliate commission shouldn't turn a bad product into a good one.
But that doesn't mean there can't be something mutually useful happening around the review.
I'm still figuring out where that line should be.
What surprised me most is that talking to the companies I review hasn't made me want to criticize them less.
It's made me want to make the criticism more useful.
For anyone else running a review or affiliate site: do you keep the companies you cover at a distance, or have you found a way to work with them without letting the relationship dictate the review?
Interesting. How are you measuring whether it is working?
As company relationships increasingly influence how deeply you investigate a product, what disclosure or review-selection safeguards would preserve reader trust beyond keeping the final verdict independent?
That's a good question. I think having a consistent baseline for every review is important. The relationship might make me go further with testing or provide more detailed feedback to the company, but it shouldn't mean products without that relationship get a lower standard of testing. I'm also leaning toward being more explicit about affiliate relationships, backlinks, or any free access directly in the review. For me, transparency around what might influence the process is probably just as important as keeping the final verdict independent.
That transparency piece is worth digging into. Could take this over email sometime — where’s easiest to reach you?
What made you pick this stack over the alternatives?
Nice work shipping it. What has been the biggest challenge since launch?
Appreciate the honesty here, most people only share the wins.
This comment was deleted 15 hours ago