Cycle vs. Harvestr — the ultimate comparison

Why you need a product feedback tool

Capturing customer feedback is the key to understanding their challenges truly and shipping the best product possible.

Although simple in theory, there is much complexity in capturing such customer context effectively. Product feedback is coming from internal stakeholders and customers from multiple sources: general communication tools (like Slack, Teams, etc), CS teams and support tools (like Intercom, Zendesk, etc), Sales teams (in their CRMs such as HubSpot, Salesforce, etc), NPS surveys (Typeform, Tally, etc), app reviews (G2, Reddit, Quora, etc), social media, and much more.

Initially, such sources are limited, and using a spreadsheet or a No Code wiki like Notion works perfectly. However, as companies grow, this approach quickly becomes inefficient and time-consuming.

Most product teams then base their decisions on incomplete customer context, leading to a suboptimal product and eventually poorer engagement and retention.

What is a great feedback tool

When they scale, companies need to have a proper feedback system in place to maintain an effective feedback loop with customers without hitting on their productivity.

Great feedback systems will ensure to complete all 4 stages of the full feedback loop: collecting feedback, analyzing its insights, collaborating on implementing new features, and communicating such releases

Feedback tools are there to help you implement such feedback systems effectively.

Here below are the 8 components required for great feedback tools, along with a comparison of how Cycle and Harvestr score for each of them:

  • Feedback capture
  • AI engine
  • Insights
  • Prioritization
  • Roadmap
  • Collaboration and editing
  • Feedback loop
  • Flexibility

The 8 components of a great feedback tool

Feedback capture

If you don’t make it buttery smooth for your team to submit feedback, they won’t. An efficient feedback capture must be both frictionless and fast.

Cycle’s way of capturing feedback

  • Dropzone: we asked ourselves: "How could this be even more frictionless?"
  • We've found that the most natural way for folks to capture feedback is to take a screenshot and drop it somewhere – no thinking, no hassle, just drop it.
  • That’s why we shipped Cycle’s feedback island. From anywhere, just drag & drop a file or a screenshot and watch the island do its feedback magic. island 2.0 (1).jpg
  • There's more... Cycle generates optical character recognition (OCR) for your images & PDFs... as well as transcripts for your audio and video files.
  • Native integrations: of course, we integrate with your favorite tools too. Hubspot, Intercom, Freshdesk, Salesforce, Gong, you name it. And, if you’re using a very specific tool, our graphql API got you covered. We’ll be happy to help you with setting up a custom integration. - Cover (1).jpg
  • Recorder: Cycle comes with a built-in meeting recorder. This way you don't have to use yet another tool and create an information silo for user interviews & onboarding calls.

Harvestr’s way of capturing feedback

Harvestr also integrates with a bunch of tools. The difference lies in the depth & user-friendliness of the integrations. Since Harvestr didn't build a very rich markdown editor, their integrations have a hard time letting you capture & format customer context smoothly (images, videos, entire Slack threads, etc.).

They don’t offer anything about recording sales, CS, or user research calls.

“We tried them all, no other software beats Cycle. We redesigned our feedback loop to extract concrete customer insights in no time with the autopilot feature. It’s a no-brainer for any product team.” – Aurélien Georget (CPO at Strapi), previously on Harvestr

AI engine

Manually processing feedback is a thing of the past. Product folks don’t have time for such grunt work, and should focus on brainstorming and delivering on the best features instead.

Cycle’s AI autopilot

We believe that categorizing feedback should be on autopilot. Cycle AI will learn from your company’s context & feature set. Once an initial calibration phase is completed, your Cycle AI will scan feedback for relevant customer quotes and map them to existing features, or create new ones.

We're also building our very own "Cycle Ask", which you can use to ask any questions about your feedback data.

We're working with the very best experts in the field to build AI that blows your mind, saves days of work, and make your product org future-proof.

“Cycle is a slick AI tool that enables teams to build better products by getting smarter on what their customers want.” – Olivier Godement (head of product at OpenAI & Cycle angel investor)

How Harvestr timidly leverages AI

The company recently announced HAI (Harvestr AI), which capabilities let you generate summaries of feedback. So regarding workflows, Harvestr users still need to manually go through their feedback inbox, highlight relevant quotes and link them to the right features & customers. Their AI won’t do the categorization work for you.

As they don’t have a rich editor to document and collaborate on features, there aren’t any generative AI functionalities available to help you document your PRDs.

Their website homepage doesn’t mention AI anywhere, so it does not seem to be part of their product vision and strategic investment.


Where feedback tools often fall short is in the query and access to the insights repository. Things end up in long lists that can barely be filtered or sorted on particular tags. This makes it quite complex to surface the most important trends and truly highlight what to focus on.

Also, such insights tend to be categorized once and can’t be accessed through multiple angles. When working on a new subject, it is quite complex to get a basic understanding of it cross-insights.

Cycle’s flexible and actionable insights

Cycle offers 3 ways to highlight trends and deeply understand customer context

  • Flexible views: display insights according to your needs — apply specific filters and sorts to create your own kanban and list views
  • Custom dashboards: we’ll generate dashboards displaying breakdowns, trends and abnormalities across your feedback data to help you prioritize. Such dashboards can be filtered based on customer attributes to focus learnings on particular market segment
  • Cycle Ask: start from a topic and get an AI-generated summary of it, cross-insights. This is the ultimate discovery tool, combining both the flexibility and depth of your research.

Harvestr’s feedback counting approach

Harvestr only allows users to manually extract insights from various feedback pieces, and link them to the right feature. It’s essentially an insight repository that can be browsed with intent but doesn’t allow for highlighting major problems or uncovering new trends.

A nice view they built is their discovery view, which projects various weights and scoring properties based on the volume of linked insights (typically including revenue metrics like ARR impact and lost deals). But again, this mostly serves for RICE prioritization in the ‘known scope of initiatives’ but doesn’t truly highlight new learnings.


Although we strongly believe that feedback has more value for creating trust with customers by closing the feedback loop, feedback can still be useful for infusing feature prioritization. The volume of feedback or requests linked to a particular problem or feature is a good proxy for assessing the demand of customers. This is why most feedback tools tend to have prioritization functionalities and don’t just stop at collecting and processing such feedback.

This is also the area where there is a clear difference on how various feedback tools approach their offering.


Cycle hasn’t invested much in this area, hence prioritization features are quite basic: mostly allowing you to create custom views through filters and sorts. One key element that helps product teams make relevant decisions though is the Voice of Customer dashboard — highlighting major breakdowns and trends of problems and features, based on the volume of insights linked.

Harvestr advanced prioritization feature set

When it comes to feature prioritization, Harvestr has you covered. They’re one of the best tools out there for baking revenue metrics into your prioritization framework. Like Cycle, they integrate with revenue teams’ tool stack (like Salesforce as an example) and let you stay on top of opportunities that make the mot sense business-wise.

You can get quite deep and custom in fine-tuning your scoring formula to match your specific requirements: specify effort, ARR impact, customer base impacted, etc to drive investment decisions.

This is especially helpful for larger enterprise customers who need to ponder requests from a large user base to decide on what to build. More agile teams tend to base their investment decisions on direct and tangible opportunities.


One of the main product function’s role is to communicate effectively. As a direct extension to the prioritization modules often available in feedback tools, roadmap views have also become an important aspect of such products.

Cycle’s custom views for internal communication

Cycle’s flexible data model allows for creating flexible roadmap views that will match any product team’s hierarchy and process. Such views options are quite limited, allowing for list and kanban display along with basic filtering and sorting options. “Simplicity is the ultimate sophistication” they say, no? 😅

We do not (yet) offer a public roadmap portal for customers to browse.

As per our opinion that feedback has more value to create trust and engagement with customers than for prioritizing features, we also believe that showing your results is more important than sharing your plans. That’s why we’ve been investing heavily in our release and public changelog functionalities, more than in the roadmap ones (see below in the feedback loop section).

Harvestr’s roadmap capabilities

Roadmap views are fairly limited in Harvestr compared to similar products in the space like Productboard or Airfocus. They don’t give as much flexibility with regards to timeline views and advanced dependencies mapping. Most teams we’ve discussed with are mostly using their Discovery view (the one used for prioritization) to convey planning communication.

The tool also comes with its public portal, allowing to communicate your various roadmaps effectively with customers and any external stakeholders, and capture interest.

Collaboration and editing

The best product teams are excellent collaborators. The product should never be a “black box” to the rest of the company. Yet, this is the number one complaint we heard from many prospects: “Nobody else than the product team ever goes into our feedback tool, it’s not a collaborative space”.

Most feedback tools chose to keep their text editing features to a bare minimum, pushing product teams to turn to separate tools like Notion or Coda to do their work. This systematically results in new information silos, which is the exact problem you’re trying to solve in the first place. 🤷‍♂️

Cycle’s collaborative doc as the atomic object from day 1

We built Cycle with collaboration in mind from the beginning. Live activity indicators, @mentions, in-line comments, advanced notification preferences, cross-source feedback loops, the list goes on.

Cycle comes with a rich markdown editor that lets you edit feedback & write specs with nifty “/” commands, images, videos & embeddings. Since product folks spend a lot of time writing, you need a tool with a top-notch writing experience.


Harvestr hasn’t invested in their editor, and doesn’t seem to foster collaboration between multiple teams across the organization (like CS or Sales) — so not breaking silos between customer-facing teams and the product ones.

The tool allows for organizing and prioritizing all features (called ‘discoveries’) into specific views but isn’t meant to get you to document and collaborate on such initiatives (like writing your PRD) — so you’ll have to get this done in another tool (like Notion or Confluence). The fact that they don’t integrate with such documentation tools either means you won’t get a single source of truth for all product information.

Their website doesn’t mention such collaborative editing features, showing that it is not the direction they’re taking for the product going further.

Feedback loop

The key behind every successful feedback system is an effective feedback loop. Feedback shouldn't be seen as a prioritization tool but rather as a tool to create trust. The system breaks when there's no way to close the loop at each release.

By closing the feedback loop with customers, companies improve engagement, trust, and loyalty – ultimately leading to increased revenue. When users see their feedback being implemented, they feel acknowledged and are less likely to churn. It creates a sense of value and trust between the product and its users.

If you've ever opened a support ticket with a software company, you know how frustrating it can be to have them tell you: "This is on our roadmap," or "I'll tell the product team." 🤔 But in many companies, communication between support and product teams is broken. The support team rarely gets an update from the product team... and neither do you.

Cycle’s native feedback loop system

This changes with Cycle. When the product team marks something as “shipped”, all the related customer insights are automatically updated with the new status. Cycle’s integrations will show this to support agents who can then easily close the feedback loop with users. Without even leaving their support tool. Additionally, your team will get custom status updates notifications by Slack and/or email so they can easily stay on top of things.

Your product team is most likely being pinged regularly with questions such as "When will this get shipped?" or "Has this bug been fixed?". Now they can simple use our /Cycle-status command to get a quick update on the status fo roadmap items tied to a specific customer.

“What hooked me on Cycle is the ability to close the loop with users. With Cycle, teams can communicate with customers in the right source with the right message when they shipped something they asked. On top of making customers feel special, it leads to higher NPS and increased Net Revenue Retention.” – Alana Goyal (Managing Partner at Base Case)

Next to the 1-1 feedback loop, Cycle’s Release module allows for organizing release notes and publishing a public changelog. “Don’t share your plan. Show your results.” they say..

Harvestr and the feedback loop

Although most of the features staged are around feedback processing and prioritization (its core focus), the tool has developed some early feedback loop capabilities. Those consist mainly of internal notifications with the feedback reporters — by email, or within the tools synchronized (such as HubSpot, Intercom, etc).

Next to that, they do have a public portal to capture requests and upvotes, and share the status of newly developed features. But the risk is that it doesn’t help in creating trust with customers upon each release. Instead, it’s creating a layer of noise that is dampening your empathy and proximity to real customer context

Also, here is our take on why prioritizing based on what customers ask is the wrong way to go.

As feedback loop features are mentioned on their main website, we can expect more of it in their future investments.


There are as many ways to ship products as there are product teams.

Products that lack flexibility usually fail to meet the expectations of power users, while products that lack opinions can be confusing and hard to use. Striking the right balance between the two is critical.

Cycle’s flexible yet opinionated data model

We’re strong believers that there are as many ways to ship products as there are product teams. Yet, we do have strong opinions on how to help you do it well.

Cycle’s approach is flexible. By default, everyone has access to "feedback", "insight", "feature" & "release" objects. But we created a way for users to customize these objects to their individual needs.

Below you can see an example of a setup in which a user has created 5 different objects: "goal", "need", "feature", "dev" & "design".

There's more. You can actually customize the relationships you want to have between your custom objects. In the example above, "goals" can be broken down into "needs", which themselves can be broken down into "features". We basically allow for multi-level nesting across objects. That's our custom hierarchy feature. Definitely the most powerful aspect of our data model.

Harvestr’s rigid approach to product management

If you already made the exercise of defining what your product process, hierarchy and taxonomy should be, then you’ll get frustrated by Harvestr’s very rigid and opinionated model. Everything is a ‘discovery’ (aka a ‘feature’) and is solution-oriented — in other words will be shipped. There is no concept of “real” discovery in the sense of a customer pain or larger subject that needs to be understood and later broken down into actual features.

To conclude

Comparing the two tools Cycle and Harvester revolves around comparing two different approaches to product feedback.

If you’re looking for a tool that will essentially help you build and prioritize a roadmap based on customer feedback (volume), then a tool like Harvestr might be a great match. They’ve been digging the space for a few years and have developed most features needed for that use case.

However, if you’re looking to capture an exhaustive understanding of your customer context while not spending hours manually processing feedback, and you believe that closing the loop with all customers at every release will help you improve customer engagement and retention, then a tool like Cycle has been designed to solve exactly for that use case.