Jellyfish Software Engineering Intelligence Platform

Jellyfish gives us real visibility into engineering work and Q&A bottlenecks that Jira alone could never provide. We are about 45 engineers, including 10 QA test engineers. Our primary use case for Jellyfish is mainly to get end-to-end visibility into our SDLC. Before Jellyfish, we had Jira burndown charts. We had no idea how much time was going into new features versus bug fixes, versus keeping the lights on. Now Jellyfish connects to our Jira, GitHub, Jenkins, and Slack.

One positive impact Jellyfish has had on my organization is that it reduced unplanned work. We found that 30% of our QA time was going into production hotfix testing, which was not planned. With Jellyfish data, we created a dedicated buffer for unplanned work in sprint planning. Unplanned work dropped to 12% in three months. It has improved developer-QA collaboration and led to better sprint estimation. We started comparing estimated versus actual allocation. Now our sprint predictions are 85% accurate versus 50% earlier. There are no more last-minute release postmortems. It also helped identify bottlenecks in QA where Jellyfish showed that tickets were stuck in 'Ready for QA' status for an average of 1.8 days due to environment unavailability. We then containerized our test environment with Docker and Jenkins, cutting wait time to four hours.

The best features Jellyfish offers are Allocation and Investment View, seamless integrations, DevEX and delivery metrics, Team Health and Work Profile, Executive Reports, and justified test automation investments, improved developer and QA collaboration, and better sprint estimation.

The Allocation and Investment View stands out for me as a killer feature. It shows us exactly where engineering time is invested, such as roadmaps, bugs, infrastructure, and KTLO. This helped us balance feature work and quality work. When it comes to Executive Reports, they provide very clean dashboards that even non-technical managers can understand. There is no need to explain Jira queries anymore.

I would also say the seamless integrations are impressive. The Jira plus GitHub plus Jenkins integration took less than 30 minutes. It automatically maps commits to Jira tickets, so no manual tagging is needed.

One positive impact Jellyfish has had on my organization is that it reduced unplanned work. We found that 30% of our QA time was going into production hotfix testing, which was not planned. With Jellyfish data, we created a dedicated buffer for unplanned work in sprint planning. Unplanned work dropped to 12% in three months. It has also improved developer-QA collaboration and led to better sprint estimation. We started comparing estimated versus actual allocation. Now our sprint predictions are 85% accurate versus 50% earlier. There are no more last-minute release postmortems. It also helped identify bottlenecks in QA where Jellyfish showed that tickets were stuck in 'Ready for QA' status for an average of 1.8 days due to environment unavailability. We then containerized our test environment with Docker and Jenkins, cutting wait time to four hours.

A few areas for improvement are that the initial onboarding period needs patience. The first two to three weeks, the data looks inaccurate until it learns your Jira workflow and Git patterns. The documentation for investment categories is confusing. Additionally, the interface can be slow when you filter data for six or more months.

I would also say the pricing is on the higher side, especially for smaller teams working under a tight budget. This means it may not be suitable for startups below 20 engineers. Jellyfish should improve the mobile dashboard view as well.

I have been using Jellyfish for the past nine years, even in my previous organization.

Our stability experience with Jellyfish over the last five years has been excellent. I would rate it a nine out of ten. It is a very stable SaaS platform. We have not seen a single major outage where the platform was completely down. In about three years, I remember only two times when dashboards were slow to load for about 15 to 20 minutes, and their status page showed they were doing database maintenance. They have a status page at status.jellyfish.co, and they are very transparent.

Scalability experience with Jellyfish has been very positive. We started with 35 engineers in our department, and now we are over 50 engineers, going to 60 next quarter, and Jellyfish handled the scale without any issues.

I have had to reach out to them a couple of times, and my experience with them was great. They are quick to respond to any of our questions or disasters, and they are also solution-oriented and very professional.

We still use Jira. We did not switch from it. We switched from only Jira reporting to Jellyfish plus Jira combined.

The deployment of Jellyfish in our environment is very easy and straightforward. It is a SaaS platform, so there is no heavy infrastructure to manage from our side. We did not need to provision any servers or databases.

The configuration process experience with Jellyfish was smooth, guided, and well-supported. It is not a plug-and-play tool where you just connect and forget. You do need to configure it properly to get real value, but the Jellyfish team makes it very easy.

Before Jellyfish, we saved $15,000 per year in reporting time. Because before Jellyfish, our engineering manager, QA lead, and I spent 10 to 12 hours per week manually creating reports from Jira, GitHub, and Jenkins in Excel. Now Jellyfish automates all of it. This is approximately 40 hours per month saved across the team. If we calculate at $40 per hour engineer cost, that is around $19,200 per year saved just in reporting. We have also seen faster release cycles with a 30% improvement, a reduction in unplanned work, enhanced QA efficiency and cost savings, better resource planning, and a 40% reduction in bug leakage into production.

The deployment frequency has increased by 30%. Lead time for changes, especially commit to production, reduced by 42%. Cycle time in progress to done reduced by 28%.

Jellyfish pricing is based mainly on the number of engineering or contributors whose data is being tracked, not per viewer. If you have 45 engineers committing code, you pay for 45 contributors, but you can have unlimited managers or viewers for free. This was good for us because we had 10 managers who needed view access but do not code. The price or the plan depends on your number of engineers.

Jellyfish is a premium price tool, not a cheap tool, but for us, the return on investment justified the cost.

My advice to others evaluating Jellyfish, as a software test engineer who has used Jellyfish for about nine years and was part of the evaluation team, is to not evaluate with fake data users. Use your real Jira and Git data. Spend time on investment model configuration. Do not try to replace Jira.

If you are 20 or more engineers and tired of Jira plus Excel reporting, Jellyfish is worth every dollar. It changed how we manage engineering from gut feel to data-driven within a period of three to four months, which was a great thing. I would rate this product an eight out of ten.

Public Cloud

Amazon Web Services (AWS)

What do you like best about the product?

The AI insights built within the product takes the heavy lifting of evaluating the data off of the user. The UI is fairly straightforward with pleasing visuals throughout the product. Pulling data from our disparate systems and present a full picture across our engineering organization in one place is a big win. When I do have to deal with the support team, they have been prompt and helpful in resolving my issues.

What do you dislike about the product?

The biggest issue I have with Jellyfish is how it attributes work done by engineers as AI assisted work. It's really making a guess based on if there are AI signals the day that they do a PR which isn't really a great determination in my opinion. Not having the ability to designate static teams has also been a problem when we have teams that are support teams who's work spans multiple JIRA projects.

What problems is the product solving and how is that benefiting you?

Before I had to go to disparate systems such as JIRA, GitLab, Github, and Linear to get data to understand our engineering team's performance. Now I can go to Jellyfish to have those data and metrics aggregated under one platform to understand the teams and organization's performance.

What do you like best about the product?

Jellyfish helps me know where I stand in AI usage and lets me see the usage of my teammates as well. It's useful for understanding how much I depend on AI. I find all the features good, and the initial setup with SSO login was easy.

What do you dislike about the product?

Nothing

What problems is the product solving and how is that benefiting you?

I use Jellyfish to track AI usage, helping me understand my dependency and where I stand. It also lets me see my team's usage, which is useful.

What do you like best about the product?

the ability to quickly view my team's progress in metrics we are tracking.

What do you dislike about the product?

sometimes the interface can be clunky and dificult to navigate. I have to know where I am going to get what I need.

What problems is the product solving and how is that benefiting you?

it is giving me visibility into how my teams are performing

What do you like best about the product?

The effort-based allocation model is the standout. Jellyfish translates raw development activity into an FTE-equivalent split across investment categories (Growth, KTLO, Support) based on actual effort signals rather than ticket counts — which is critical, because counting issues badly misrepresents real work. This is what made my monthly CEO and board reporting credible: leadership could trust an "X FTE went to Growth this month" figure because it reflected effort, not just Jira volume. The mapping of investment categories to our own strategy taxonomy and the month-over-month trend views also let me show direction over time with minimal manual assembly, and the delivery and AI-assisted PR metrics fed our AI-adoption reporting well.

What do you dislike about the product?

The biggest frustration is the newer AI impact reports: there's no way to export the underlying data to CSV. The reports render fine in the UI, but if you need the numbers for your own analysis — cohort breakdowns, month-over-month trends, feeding an executive deck — you're stuck. You can look at the data but you can't take it with you, so getting it into a usable format means manual extraction rather than a simple download. For a metrics platform whose whole value is helping you report, locking the data inside the dashboard defeats the purpose.
On top of that, I've seen the per-tool detail figures and their aggregate rollups diverge by around 10 percentage points, which is a real problem when the aggregate is the number going in front of the board. Both issues are workable, but they add reconciliation and extraction steps to what should be a clean pull.

What problems is the product solving and how is that benefiting you?

It answers the core question my leadership asks every month: where is engineering effort actually going, and is it aligned to strategy? The FTE allocation across Growth/KTLO/Support gives me a defensible, repeatable way to report that, plus the delivery and AI-assisted-development metrics I need to show whether our tooling investment is moving velocity. The benefit is a monthly executive narrative grounded in effort data rather than anecdote or raw counts.