60 Day Plan Look-back

This past month we completed our 60 day plan, and I wanted to do a quick retrospective on all the accomplishments.

We've installed an advanced monitoring system that gives us real time logs of all aspects of our system, which previously we were not seeing. Now I can log into our system at any time and scan for any recent errors, which is a huge improvement in the visibility of errors over the previous situation.

We've updated the userland UI. While for me UI is never complete, we've made some serious improvements, especially regarding onboarding and forms- we've solved some serious compatibility issues with major products as part of this. Some parts of our system were built with a little bit of haste as can be the case with startups so a lot of this process was simply going through our product and noticing every little inconsistency or problem and correcting. Consequently whenever I am watching someone demo our product now I breathe a lot easier knowing we've gone through it very thoroughly and on every imaginable system and seen all the bugs.

We've added several features too, the most recent of which I am very excited about. We created a list of companies with our data and with Owlers data and were able to see some incredible insights just by combining these two sets. With a little more work hope to roll this into a product in the future.

Data handling is probably the most important item in our 60 day plan. I went through our entire data pipeline and put in all sort of diagnostics to discover errors that before we were only vaguely aware of. We knew something was wrong, because we'd see problems, but now we know with certainty when something goes wrong, what is going wrong, and usually how to correct it. We made a few very important changes to the process logic after examining all of these diagnostics and saw a huge improvement in the new data we ran through it. Right now we are in the process of re-running all past data, and my expectation is that that data on our site will approach an extremely high level of accuracy once complete. Our data as is is pretty accurate to a point but occasionally we'd run into things missing or inconsistencies so its as though now we are going from say 88% to 99% accuracy in the very near future as a result of this work. While it may not seem like a great difference I think it is a huge difference in reality. If someone sees that our data is wrong they are not going to trust us as a platform, even if that data being wrong is not symptomatic of our other capabilities outside of the data.

We've made changes to the UI to make it more responsive and app-like. We discovered that some aspects of the site simply would not work on mobile and correcting this was a high priority whether it be styling changes (looks bad on mobile but does work) or more (it is dysfunctional on mobile in some way or simply will not work). This was especially true for the onboarding process. Some parts of it just looked awful on mobile or were extremely hard to deal with for one reason or another.

Right now my focus is re-running old data and then on new improvements to the profiles and new products.

60 day plan update / post login screen

We're about halfway through our 60 day sprint and we've accomplished a lot of unsexy things that we'd been putting off - like setting up a complete clone of our production environment, setting up fully functional logging on all of our systems, and debugging our data pipeline.

While these don't make for great line items to put on our changelog, they are absolutely critical system improvements that had to be done immediately for site stability, and development ease.

For instance, without a clone of our production environment we don't really know how code will work until it is live at our url, and that's somewhat dangerous. Before we had this setup, we'd occasionally experience unexpected outages because of unforeseen differences in our dev/prod/local setups. That's just one example, but there are many. My nightmare is seeing "Internal Server Error" instead of our homepage, so I'm very happy we have all this in place now. It also makes testing things a lot safer, and easier.

We'd known that our data pipeline would occasionally have errors which would create problems in our data. Although most of these problems were isolated to a single profile, we take data accuracy very seriously, so that's been a constant focus in the past 30 days. I'm basically taking every case where we see errors and following it through our system trying to find bugs. I've discovered maybe 3-4 problems, but we're not done and I won't rest until we see even isolated errors go away.

Installing a better logging setup has greatly helped diagnose previously unknown errors.

Today we've just added a new post login menu that will make it easier to navigate our site for logging in users. Before the site would take people to the profile edit page, for convenience, but this new menu allows a user to easily see some of the most important things to do on 202Works.

My schedule this week is crammed with features instead of debugging and improvements, so expect another update soon.

Magic Links, bug fixes, improved validation experience, 60 day plan

This week we moved into our new offices (yay) and we're very excited to be back at it. We rolled out an update to improve the login experience for onboarding users, by providing them with Slack-like magic links when coming to our site for the first time, this is an improvement of our previous system of tokenized urls, which would error out after 1 attempt (for security). This means people can check out our onboard experience and comeback to it from the same link without getting locked out, and needing a password reset

During the course of that upgrade we uncovered some new bugs in the login and invite system which we were able to fix, including improving the form validation experience.

It's part of our plan to improve error handling in general now that we are moving past the MVP phase of our product. Once we raised funding we crafted a 60 day timeline to improve the polish and stability of the site in general. These tasks include preparing to scale, adding a more fully featured monitoring system, as well as updating the userland UI, data handling improvements, adding several features, and making the site more responsive and app-like.

Onboarding, Stripe Connect, error corrections

With the new year approaching we've deployed two critical updates that push us forward as a company.

The first is our new onboarding flow. We've spent some time perfecting it across mobile and desktop platforms. While undoubtedly some changes will occur from here, it feels good to have a significant upgrade to our onboarding experience be live.

The second update is our integration with Stripe Connect. Companies and individuals are now able connect with 202works Marketplace to receive and manage payments through our platform.

As part of this major deployment several bugs were corrected that some of you may have noticed if you've been using our site regularly.

Added Alert Integration

For some time we've been trying to find a nice solution to let us know when user began interacting with our site. While our vendor Intercom is supposed to support slack alerts for their Leads API, from our testing using our site and Postman, we just haven't seen that happen.

Fortunately, while I was taking a break from working on onboarding over the weekend I found a perfect integration that works well and will alert us when we get an invite. I still want to test it and make sure it does not have any of the gaps that I saw with the out of the box api/slack integration, but so far it looks really good.

This is important because our target audience are very busy people and if they are having trouble signing up or forgetting to go through our process we want to know immediately. We can spend 1000s of hours building in ease of use, but it is priceless to know when someone is stranded without them telling us, among many other use cases.

New Onboarding flow being tested

So for the past 2 weeks we've been working on finalizing a new onboarding flow that will have our experts claiming data and uploading a photo as part of their onboarding. This is a major step forward. We're trying to mimic the onboarding flow of large sites like LinkedIn, down to the fit and finish.

The new flow has been up for about a week on our test server and we've been adding changes to it. I'm hoping to have it on production this week. Because users are so important to us, we are taking every precaution in testing this new flow fully before putting it in production.

Security Updates

This week we were able to resolve some holes in our site security. While this is always a touchy subject to discuss we are proud in our pursuit of an ever more secure service.

Previously it had been possible to access our data through a targeted attack and steal it, but we closed that particular hole this week, and are working on a solution to prevent such vulnerabilities in the future.

No site is invulnerable to attack but it is important to not make it easy.

Templated Password Resets / Changelog!

We're adding a change log so that you can keep up with everything we're doing on the development side. This week we're getting ready to roll out a bunch of updates to the site functionality, but so far we have rolled out templates for password resets.

We're starting a changelog

Big news today, we're starting a public changelog so you're always up to date with all the updates, improvements and fixes that are made in .

Even though we work on all the time, sometimes it may seem that not much is happening. This changelog is here to improve that very important part of the communication between you and us.

You'll always receive an update when we change something in the widget, or if needed all things we've changed is available on our public changelog page.

No published changelogs yet.

Surely 202works will start publishing changelogs very soon.

Check out our other public changelogs: Buffer, Mention, Respond by Buffer, JSFiddle, Olark, Droplr, Piwik Pro, Prott, Ustream, ViralSweep, StartupThreads, Userlike, Unixstickers, Survicate, Envoy, Gmelius, CodeTree