This is the second in my series on the 3 Rs of AI Search Optimization. If you missed the last one, click here to read the first post in the series.
If you’ve ever met me, you’ve probably realized that I’m an optimist.
I’m very glass-half-full. I come by it naturally.
Most of the time, it’s an asset. But some of the time it’s a problem. If I can’t find the “bright side” of something, I’ll avoid it altogether.
Which is why this post is coming to you later than planned. There is just no way to spin into a good thing the fact that (because AI) we all need to become brand reputation managers.
Frankly, it kind of sucks. But it’s above my pay grade to fix the underlying problem, so … it is what it is.
The Problem of Brand Reputation in AI Search
Imagine your next customer. She’s pretty sure she’s going to hire you, but wants to do her due diligence, to make sure your company is legit.
She types in “[brand name] reviews” into Google.
Or “Is [brand name] a trustworthy company?” into ChatGPT.
What she finds determines whether or not you get hired.
Before AI search, it was pretty easy for you (and for your prospective clients) to know if you had a good reputation or a bad reputation online. It was all there, clear as day, in the star ratings on your owned business profiles—on Google Business, Facebook, LinkedIn, TrustPilot, etc.
With AI search, things have changed.
First of all, AI engines don’t just look at “official” reviews of your company posted online on your owned profiles. It looks at reviews everywhere.
Now even one lukewarm review (e.g. ”my experience was just OK”) of your brand hidden in a corner of the internet is likely to get reproduced (word-for-word, with a citation link) when someone comes to AI with a question like “Is [brand name] worth it?” or “Should I hire [brand name] for [service]?”
Why do negative and somewhat-negative reviews get so much attention in the answers to these kinds of prompts?
Because AI is the great equivocator. It’s noncommittal by design.
Across all the generative models I’ve checked, the prompts asking for reviews of a company almost always start by summarizing positive reviews and then ends with one or two negative reviews.
It’s perhaps most obvious in Amazon:
(Self-contradictory proclamations don’t phase AI one bit. In the Amazon summary above, notice “connectivity” is both a pro and a con.)
I’ve rarely seen any AI models present strong positive reviews without balancing it out with at least one negative one.
Even if your brand has a flood of positive reviews, just one negative review is going to get a lot more visibility than it deserves.
AI can pull reviews from just about anywhere on the web, but one source that almost always gets queried is Reddit.
This makes Reddit ground zero for controlling your brand’s online reputation.
I recently found F5BOT, a handy free tool for setting up Reddit alerts. Just give it your brand name and your personal name (and any other keyword you’re interested in tracking) and you’ll start getting alert emails within minutes of somebody publishing something on Reddit with that word or phrase in it.
Like it or not, we’re all in the reputation management business now.
Setup F5BOT to stay up to date on what people are saying about your brand online, and start regularly checking the output for prompts like:
- is [brand name] worth it?
- is [brand name] trustworthy?
- [brand name] reviews
- I’m a [target persona] and I’m thinking about buying [service/product] from [brand name]. Is that a good idea? What kind of results can I expect?
Notice I didn’t even get into fuzzy search, look-alike brands, and the very common problem of other brands’ negative reviews showing up in the “negative review” slot of your results.
You can read more about that in my Cheat Sheet to Showing up Well in AI Search Answers. I don’t have a magic solution to that problem. If I ever find one, I’ll tell you!
