Changelog

Todo -> Doing -> Done. But what happens when a feature is marked as Done? It’s deployed and live with the customers. That’s undoubtedly the most crucial time in the feature cycle, and that’s exactly where existing workflows stop today.

It’s nuts!

At Bucket, we want to add a final step to the feature delivery workflow: The Evaluation step. In the Evaluation step product teams make sure that their customers actually like the feature they’ve released. We believe most features should go through this step.

We’ve come to expect almost every other step in the feature delivery workflow, like testing, deploying to production etc, to be automated. We believe that the evaluation of features should also be as automated as possible.

Last week we announced how you can plan features on Bucket. The Planned state is for when the feature is in development.

Today, we’re announcing the remaining states: Evaluating and Done. Together, they make up a complete, automated feature evaluation workflow.

Automated evaluation

When you log in today, you’ll see these columns: Planned -> Evaluating -> Done.

When a Planned feature is deployed (and marked as Done in JIRA or Linear), it automatically goes into Evaluating mode on Bucket.

Bucket reports engagement metrics on all features in Evaluating mode to Slack every Monday.

When you have enough insights to decide if the feature has reached the impact you expected from it, you mark it as Done on Bucket. If the feature needs more work, you rinse and repeat the workflow.

Release date

We’ve also shipped the ability to set a Release date for each feature. You can set it manually or let Bucket do it. Once data for a feature starts flowing in, the feature changes state automatically from Planned to Evaluating, and Bucket will automatically set the release date to the current date if a date isn’t already set. You can modify the release date later, if needed.

The release date will also appear in the Slack reports, so you can easily track how long a feature has been in evaluation.

After 2-4 weeks, you should start to see leading indicators on adoption and churn, which will indicate feature success or not. If you need to dive deeper and speak with customers, Bucket makes that very easy, too.

Happy shipping!

Adding tracking to a feature release is often an afterthought or forgotten altogether. Many engineers have received a DM from a PM around the time of feature release saying something frantic like:

‍“did you remember to add tracking? we need to track this feature!”

We believe there’s a better workflow for this, and that workflow unlocks some very useful automations.

Today, we’re introducing Planned features on Bucket!

When a feature goes on the roadmap, there’s often a story issue created for it in the issue tracking of choice. It describes the customer problem, maybe a potential solution and ideally also a basic tracking implementation specification (tracking spec). The tracking spec describes which instrumentation to add to the code, so it can be measured and evaluated post-release.

Now, you can plan features on Bucket and hand them off to engineering as a ticket that includes the exact tracking spec - right from the Bucket UI.

Here’s how it works.

First, track a new feature. In the new feature modal, you can now enter events and/or attributes that don't exist yet. In this example, we want to fire off the “Pinned card” event whenever a customer pins a card in our product UI.

Once created, you now have a feature tracking spec that you can share with your engineering team. It contains all the information needed for the engineers to know exactly what event or attribute to track before releasing this feature.

You can share the feature tracking spec as a link, copy as Markdown or create a Linear issue straight from Bucket. You can also add a note about when to fire the event.

With the feature created on Bucket, the system will listen for events with the name “PinnedCard”. When engineering is done with the feature and deploys it, Bucket will get notified as soon as a user interacts with the feature. If you want to, you can get a Slack notification when that happens.

Besides getting the feature tracking spec out of DMs and into the issue tracking system, there’s another huge benefit to creating planned features: Automated feature evaluation! 🥁  

‍

Automated evaluation

We want to empower product teams to make impactful features. To validate impact, you need to track customer engagement. If customers aren’t adopting the feature or are churning away fast, you need to act on it.

With Planned features, Bucket will know when a new feature is released and can therefore automatically start to report feature engagement metrics to your team. By default, Bucket will send you a feature engagement report to Slack every Monday – empowering everyone in the product team (engineers, PMs and designers) to act on bad engagement or adoption.

With the Bucket insights, the product team can quickly determine if the feature is “Done for now” or needs another iteration.

If you need to dig deeper and ask for customer feedback, Bucket makes it easy to figure out which customers that are relevant to reach out to. Here’s an example of that:

That’s our workflow: Repeatable and automated. Just how we like it!

More updates soon.

If you’re detecting high churn or low activation on a key feature, you want to reach out to the relevant accounts for feedback, and learn how you might improve the feature.

We’ve now made this workflow easier on Bucket.

When you dive into an account, you can now filter by feature usage and see a list of users that have interacted with the respective feature. The users are sorted by feature interaction, so you can quickly pick the most relevant user to ask for feedback. From there, click to send the user an email.

Here’s what it looks like:

Make sure to send the reserved "name" and "email" user attributes when tracking users to see the name and email link in the Bucket interface. For example:

bucket.user(“<userId>”, {
  name: “John Doe”,
  email: “john@doe.com”
}

To see all users of a company, remove any feature filter in the top right when looking at the company page.

More updates soon.

We’ve just shipped a highly requested feature: Saved segments.

This feature allows you to group companies into reusable segments, which you can apply to any feature report. You group companies by company attributes.

Here’s some examples of common use cases for segments in B2B SaaS:

1. Active
   Find your active accounts by filtering down to accounts that have logged in recently. You may also want to add an activation threshold, so you get active companies that have also completed onboarding. For example, "completed_onboarded" is "true".
   These are the base accounts you want to measure feature retention and churn against.
   
   
2. Customers
   Create a segment just for your paying customers. You can do this by either filtering on a "plan" attribute or check if "monthlySpend" is greater than "0". We recommend the latter as it is immune to plan name changes.
   ‍
3. ‍‍Exclude staff
   ‍Filter out internal interactions by setting a filter like "name" is not "MyCompany".

Applying a segment to a feature report

When you’ve defined your segments, you can apply them to any feature report. This means the feature report data will be filtered to only include the companies that match the chosen segment.

To apply a saved segment, select it in the dropdown when tracking a new feature. Like so:

You can of course change this later. The default segment is the built-in Active segment.

Migration note to existing users: This feature replaces the individual target audience filters on features. Any such filters have been converted into segments. You can rename those segments when you log in.

Happy shipping!

Dashboards always come with a trust issue: Can you trust the metrics? Is the tracking still working?

To address this issue, the Bucket feature report now includes tracking health indicators to help ensure that the data is still coming through and that the report is based on the most recent data.

For event-based features, you’ll now see the following indicators in the sidebar. You’ll see when the event (including any attributes) was first and last seen. You’ll also see a count for the past 7 days.

If there are any events that aren't associated with a company, you’ll see the red-colored warning (screenshot to the right) since those events aren’t included in the metrics until they get associated. You can learn more about associating user events with companies in this help article.

For attribute-based features, you’ll see when there was first a match on the feature’s adoption criteria for any company. You’ll also see when there was last a change to or from the feature’s adoption criteria by any company.

Happy shipping!

‍

You can now customize your organization with your own logo. It’ll show in the top left, next to the app switcher. Simply drag and drop to set it up.

Here’s how:

More updates soon!

We’ve recently released Focus Mode and Feature Views. Today, we’ve fully integrated the two which enables you to set up Focus Slack reports per Feature View!

As your account grows on Bucket, having all features in the same place gets messy. Therefore, you want to use Views to group and organize features by product team or product area.

With today’s release, you can enable a Focus report per each View. This means that “Team Mobile” can get a dedicated report of their features to “#product-mobile” and “Team Payments” can get their weekly report to “#core-payments”. And so on.

Focus and Views works out of the box but is designed to be flexible enough to fit any organization as it scales. Once you have multiple Views on Bucket, you may want to disable the Focus report on the “All”-view, so you don’t receive duplicate feature reports.

Happy shipping!

‍

Large accounts quickly track a lot of features, which means a growing need to get those features organized. A common way to group features is by product team or by product area.

To support this, we’ve just released Feature Views!

Feature views are highly flexible, so they fit any organization. For example, you can create a view for all features that are owned by “Team Mobile”, or a view for all features related to “Payments”. You can navigate views from the updated sidebar.

Whenever you track a new feature on Bucket, you can choose an view for it.

Here’s what it looks like:

More updates soon!

‍

We’ve just updated our UI with a brand new sidebar, and a bunch of general UI polish to ensure the app stays clean and keeps focus on the content.

In the updated sidebar, you’ll now see the app selector at the very top. We’ve also moved the “Track feature” button to the sidebar. This frees up space in the top right of the screen which is a location we’ll generally use for actions related to the current view. 

We’ve also streamlined common UI components like buttons and inputs for better consistency, and improved focus/keyboard navigation. Oh, and dark mode is now a bit more easy on the eyes.

Here’s what the updated UI looks like:

Besides the facelift, the updated sidebar unlocks some great features in the pipeline… 👀

More updates soon!

We’re delivering engagement metrics for any feature with Bucket’s turn-key feature report. It empowers product teams to quickly learn, if their feature is done or needs another iteration.

Having feature-scoped analytics at your fingertip is a major step in the right direction compared to the status quo. However, the problem with dashboards is this: They’re only useful when you look at them.

We believe that product teams need feature metrics to be pushed into their existing workflow for them to be used sufficiently and repeatedly.

Today, we’re releasing the first of several workflow features that’s designed to embed Bucket’s feature metrics into the existing workflow of modern product teams.

But first, let’s quickly unpack the problem.

Short feature attention span

Nowadays, product teams are really efficient in designing, building and delivering features. Once the feature has been deployed, the team (rightly) celebrates, and soon after moves on to the next backlog item. That’s the problem.

The existing feature development toolkit has improved dramatically in the past decade. As an industry, we now have streamlined, powerful tools that help us “ship faster”. It’s terrific for releasing features that work technically.

However, once the feature is deployed and is with the customers - which objectively is the most important part of the cycle - our existing tools takes us no further and we drop our attention at the most crucial time. It’s pretty nuts, really!

Introducing Focus Mode

Focus Mode enables product teams to keep an eye on the features that need attention the most - often newly released features - by pinning those in the UI and reporting their metrics to Slack.

Here's what it looks like:

Once a feature goes into Focus, it’s pinned at the top of the redesigned list view.

More importantly, it’s also featured in the new Focus report that’s sent to Slack every Monday morning.

The purpose of the report is to enable teams to easily skim live engagement metrics for newly-released features at least once a week. With this data, product teams can follow the “health” of a feature as it goes into production to the customers.

We believe such a workflow is transformative for any product and engineering team in terms of prioritizing backlog items vs live features that aren’t doing very well.

Rinse and repeat

How long should features stay in Focus, you might ask. In short, until the team can decide on one of the following three outcomes:

• Validated
• Needs new iteration
• Remove it

Most features will take at least a month to evaluate: Customers need to become aware of the feature, try it, become active users of it and hopefully stay retained over time.

Once the product team has sufficient data from the Bucket reports, and potentially customer feedback, too, they can make a decision on the feature and take it out of Focus.

Then, rinse and repeat with the next feature release.

Happy shipping!

‍