In reworking my website over the past few weeks, I’ve had a lot of time to stare at my writing, in particular the reviews of films, books, music and games. I realised, quite quickly, there was something really lacking in that department. I never claimed to be a seasoned critic of any kind, but even I could see that the reviews had gotten vague and lazy. It’s all very well for me to be using this website as a memory extension – did I read this, and did I like it – but if I’m writing 200 words about something, they might as well add something to the conversation.

Feedback on the feedback

I used Claude AI as a sounding board to workshop where I was going wrong – fed in a few of the existing reviews and prepared myself for the feedback. Anyone who thinks AI just tells you what you want to hear and feeds your ego... I beg to differ.

There were three key findings:

• I was generally too vague. There were elements of my reviews that could be written about anything. My Hogwarts Legacy review has half the word count listing stuff that could be found on a shop page for the title. I loved that game, but most of what I’d written was straight from the suppliers.

• Naming the category instead of the instance. In my recent Emily Henry review I named three great things… about romance books in general. The well-worn path of friends to lovers, the split narrative, the will-they-won’t-they. But what specifically about this book? We’ll never know.

• The more I like something, the more vague the review. This was a bit of a surprise but was true across the board. If I have a criticism, it’s easier to grab on to that and write about it. But praise is harder – I like it because I like it. As mentioned above, Hogwarts Legacy was a wishy-washy review despite loving the game. This was a startling finding on a recent Roxette post: ‘there are some real gems on here’ without naming a single one of them.

Asking the question

With a view to what was going wrong, I worked on a structure and method for the reviews that would work as prompts going forward. In creating a structure, it helps keep the blank page from putting me off and gives me key questions to start from. For all four types of review, it’s the same idea:

1. Why this title?
2. What specifically about it did you like?
3. Why might other people like it or not?
4. Why the particular score you’ve given?

I’m not going to lie, it’s been a slightly torturous process. I could write and write and write sentences all day long but to actually have to give opinions and tie down those specific thoughts, I don’t know why but I find it really difficult to do.

And the structure has been through a lot of revisions. I briefly tried recommending similar types of content but found that detrimental rather than motivational. And I’ve been getting my head around the best way to write a 'why this score' section – a 4 out of 5 isn’t always a top score just missing something. Sometimes, it just doesn’t feel like a 5.

Then and now

It’s been worth the effort though. Recent reviews on my website are night and day better than ones that have gone before. It’s not just that word counts have crept up, that wasn’t the point of the exercise, but they have substance and a reason for existing. In ten years when I look back at an album, it’s not going to say there’s some gems on there that I then have to search for. It’s going to say what the good songs are and why.

It’s not my intention to go back and re-review everything on the site and write it up better, although that was my natural instinct once I could see just how lazy I’d gotten. But if I do revisit something, then I’m hoping the second review will add a lot more to the story. I did it recently with Crocodile Shoes and you can see the change quite neatly, with two reviews written about two months apart.

Before:

It's a great country album, that distinctive gravelly voice across some cowboy dreaming songs. I really liked it.

After:

The opening track, Crocodile Shoes, is pure cowboy and it put me in mind of Wand’rin’ Star as sung by Lee Marvin in Paint Your Wagon. I suspect that’s why I mentioned a gravelly voice in my first review. But that’s not what this is - it’s an unexpected voice. You know that Geordie bricklayer from Auf Wiedersehen, Pet? Well, turns out, he has an incredible and emotive vocal, losing the accent, bringing an earnestness and softness that feels like it would take years to develop.

The slower path

I’ve also noticed that it makes me consume the thing differently. I’m paying more attention to the albums as I listen. I’m looking out for quotes in the books I’m reading. For films, the live blog was already doing half the hard work for me, but I rarely scrolled back up and gave it a read to summarise in the review.

I had a vague idea that I was veering towards completing things quicker and quicker, maybe just for the sake of logging them on the site. Obviously, it’s my aim to read, play, listen and watch everything that’s ever been created... but this way I’m certain I’m appreciating what I’m consuming now rather than just ticking it off and moving on.

This has been an eye-opening and quite sobering process, but if I ever get too big-headed about it, I now have four documents saved down that detail just how bad my writing had become. Meanwhile, I’m off to replay Hogwarts Legacy and figure out what it is I like about it, more than just what I’ve cribbed from the publishers.