Are we having fun yet?

Things break when the action becomes the goal.

I can spend hours reading reviews on Google Maps. Coffee shops, doctors’ offices, post offices, parking spaces, anything goes. Google Maps reviews are automatically shown sorted by relevance, and app users can sort the reviews by recency, highest first, or lowest first. Of course I click on “lowest first” and read what others find worth complaining about. A common category of negative reviews is complaining about something that the listing owner can do nothing about, like construction in progress or no public parking on site. Some will use a listing to comment on the nearby place that may not have a listing on Google Maps, for example, complaining about the volume or quality of music choices of the nearby property. A recent 1-star review I saw blamed the coffee shop for their rule enforcement. The rule was “no laptops on weekends,” which the commenter found understandable when the place was busy, but not when it was empty, which is how it was when they visited, and they were still told they couldn’t use their laptop. You and I might have different opinions on what’s “right” here. This is part of the fun of reading reviews, that pretty much nothing is black and white.

The replies are also very, very good. A restaurant owner lovingly asked for patience from a reviewer who lamented the slow service and getting the wrong order. See, that waiter was the owner’s son and he was new in the job, learning, and everyone needs words of encouragement from time to time, but especially someone just starting out in a new field. There are many screenshots of reviews and replies on my phone. It’s a free, simple way to get to know a place and what the managers are like without ever going. Most listing managers try to solve the issue and win back the customer. But all kinds of flavors go for review replies: some alleging that the negative review is from a competitor, some get defensive and blame the reviewer for doing something wrong, some request that you reach out to them via another method to clear things out, some are ironic, some will do scene-by-scene breakdowns of what actually happened, and some threaten to take legal action if the review isn’t removed.

I’ve noticed the blanket statement-ification of review replies and have only ChatGPT et al to blame. The replies have lost their value and meaning. The meaning is lost in multiple paragraphs offering a mumble of words about service and client satisfaction that feels lacking. These replies are virtually worthless to me as a potential customer because after reading them, I still have no idea what type of treatment I should expect. I can’t tell what’s important to them or what to expect in an unpleasant situation, like the time the waiter charged me 41 euros for a 14-euro meal and claimed I had tipped him the rest. (I hadn’t.)

Google Maps review replies are one of the victims of AI, many of which have been killed in the name of speed. I saw someone promote their upcoming AI webinar by saying they had built a website in 52 seconds.

I laughed.

Everyone’s asking everyone, especially those in tech-adjacent spaces, how AI’s affecting their work and what they think the future will look like. Everyone’s so serious about it.

But that’s what I did when I read the 52-second website message, I giggled.

It’s okay to say things are silly when they are.

The blanket statement-ification of review replies is silly because the time efficiency gains or cost savings don’t make up for what’s lost, and what’s lost is the understanding a potential customer gets of how they will be treated, especially if they end up dissatisfied. Which is the entire point of having the Maps communication channel in the first place.

The 52-second website promo message told me nothing about why it was built, it was all about how. We’ve come to this point where the action becomes the goal in itself, replying to a review or building a website, and that’s silly.

The disguise of work isn’t work, and more importantly, it doesn’t work.

AI Policy: I personally write each draft and final copy on this website. All content reflects my own thinking, ideas, style, and craft. I do not use AI such as ChatGPT to generate articles. Occasionally, I ask AI such as Formalizer or Claude to summarize or restate my own ideas and may restructure sections based on the response.

Delfina Hoxha

Author

I’m Delfina Hoxha, the founder of Little Language Models, an information architecture consultancy in Vienna. I’ve influenced big and small decisions that led to exceptional user experiences for universities, libraries, and global tech brands. Follow to not miss upcoming weekly IA insights.