Am I the only one who finds Writesonic's outputs too verbose?

I’ve been conducting a series of side-by-side comparisons for our internal documentation team, and one tool consistently sticks out for the wrong reason: Writesonic. The claim of "high-quality, human-like content" seems to translate, in practice, to "use three sentences where one would suffice, and drown every substantive point in a sea of adjectives."

Take a simple, real-world brief we used across several tools. The goal was to generate a concise, two-sentence summary of a new Kubernetes cost-optimization feature for a changelog.

**The Prompt:**
```
Tool: Generate a brief, clear changelog entry.
Audience: DevOps engineers.
Key Info: New feature "Node Sleep Mode" for our K8s cost-optimization tool. It automatically scales down underutilized worker nodes during off-peak hours, reducing cloud costs. It's now in beta.
Tone: Technical, straightforward, no marketing fluff.
```

**Writesonic's Output:**
```
We are absolutely thrilled to announce the exciting beta release of our groundbreaking new "Node Sleep Mode" feature, designed to revolutionize your Kubernetes cost management! This innovative and intelligent capability proactively identifies underutilized worker nodes during your predictable off-peak periods and automatically scales them down, thereby orchestrating significant reductions in your overall cloud infrastructure expenditure and empowering your FinOps initiatives.
```

**Claude 3 Opus's Output (for comparison):**
```
The beta release of "Node Sleep Mode" introduces automated scaling of underutilized worker nodes during off-peak hours. This feature aims to reduce Kubernetes cluster costs without manual intervention.
```

The difference is almost comical. Writesonic's version is 58 words. Claude's is 31. But it's not just about word count; it's about signal-to-noise ratio.

Writesonic's output is riddled with empty intensifiers ("absolutely thrilled," "exciting," "groundbreaking," "revolutionize," "innovative and intelligent," "significant") that a technical audience immediately discards. It also commits the cardinal sin of telling the reader what the feature is "designed to" do, instead of just stating what it does. The final clause about "empowering your FinOps initiatives" is pure marketing glossary bait, adding zero informational value for an engineer scanning a changelog.

To get something usable from Writesonic, I had to edit aggressively:
1. Strip all adjectives and adverbs of enthusiasm.
2. Convert passive, marketing-focused phrasing ("designed to revolutionize") into active voice ("reduces").
3. Delete the entire final clause.
4. Re-structure the remaining core facts into a logical order.

The end result after my edits was essentially what Claude produced natively. This pattern repeats. Whether it's a data pipeline setup guide or a serverless architecture overview, Writesonic defaults to a verbose, padded style that assumes the reader needs to be *sold to*, rather than *informed*. It adds layers of "context" and "benefit framing" that are obvious to the target audience or, worse, actively distracting.

Is this a case of its training data being heavy on SEO blogspam and corporate press releases? Or is there a configuration buried deep that turns off the verbal inflation? I've tried instructing it to be "concise" or "succinct," but the underlying tendency to amplify and decorate remains stubbornly present.

For a community dedicated to clear comparisons: has anyone else found this to be the case? More importantly, has anyone discovered a prompt technique or setting that forces Writesonic to behave like a tool for informed adults rather than a used-car salesman?

-- Cam