# UI in annotation covering up the space

**URL:** <https://support.prodi.gy/t/ui-in-annotation-covering-up-the-space/7600>\
**Category:** Uncategorized\
**Tags:** usage, front-end, spancat\
**Created:** [August 14, 2025, 10:37pm UTC](https://support.prodi.gy/t/ui-in-annotation-covering-up-the-space/7600 "2025-08-14T22:37:22Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![hazelkang](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/hazelkang/32/4627_2.png) [@hazelkang](https://support.prodi.gy/u/hazelkang)\
**Post date:** [August 14, 2025, 10:37pm UTC](https://support.prodi.gy/t/ui-in-annotation-covering-up-the-space/7600/1 "2025-08-14T22:37:22Z")

</div>

I have multiple span classifications I’m working on and it is long. So I can’t see the text that I try to annotate at all. Could this be fixed?

 ![image](https://us1.discourse-cdn.com/flex020/uploads/prodigy/original/2X/2/2646cf6f5aea2dea5aabc1d9800a1564841dbbc8.png)

---

<div class="post-metadata">

**Author:** ![ines](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ines/32/3_2.png) [@ines](https://support.prodi.gy/u/ines)\
**Post date:** [August 19, 2025, 7:39am UTC](https://support.prodi.gy/t/ui-in-annotation-covering-up-the-space/7600/2 "2025-08-19T07:39:16Z")

</div>

Hi! A similar question came up in this thread a while ago, so re-sharing the response with solutions:

> [@Number of Labels](https://support.prodi.gy/t/number-of-labels/4405/2):
>
> Hi! The label bar at the top has the class `.prodigy-title`, so you could do something like this via the `global_css` config setting to give the container a maximum height and make it scrollable:
> 
> ```css
> .prodigy-title {
> max-height: 150px; /* or any other height */
> overflow-y: auto;
> }
> 
> ```
> 
> That said, if your goal is to annotate named entities or similar spans and you ended up with this many labels, we'd typically recommend rethinking your label scheme and structuring your task differently. You're making your life a lot harder this way, and it'll be much more difficult to create consistently annotated data with enough coverage, and your model will be much less likely to learn from it effectively. See this thread for more background and suggestions:
> 
> [NER with dozens of entities - #3 by ines](https://support.prodi.gy/t/ner-with-dozens-of-entities/4139/3)

Given your labels and complexity, I think it could be helpful to break down the task and focus on a subset at a time, e.g. `SENTIMENT` or `EXTERNAL`, and combine the data later on. This means the UI can be simpler to navigate and the human annotator doesn't have to mentally "iterate" over all categories for each example ([also illustrated here](https://speakerdeck.com/honnibal/practical-tips-for-bootstrapping-information-extraction-pipelines?slide=80)).

If you haven't seen it yet, this section in our case study with S&P Global illustrates a similar approach, making the overall annotation 10 times faster as a result: [How S&P Global is making markets more transparent with NLP, spaCy and Prodigy · Explosion](https://explosion.ai/blog/sp-global-commodities#workflow)

 ![sp_prodigy-comparison](https://us1.discourse-cdn.com/flex020/uploads/prodigy/original/2X/4/44c8608bc37226bfd3b613ff3336003e68676267.jpeg)

> However, the cognitive load from having to consider this many attributes at the same time made the process incredibly tedious and too slow to be practical. [...] So the team tried something else: focusing on a single label at a time and making multiple passes over the heards data, once per attribute. Although this sounded like more work at first, it drastically sped up annotation time by over 10×.