10 tips that helped our B2B SaaS product score a 60+ NPS in the midst of a pandemic

Amay Jhaveri
14 min readNov 26, 2020

When I tried starting my first company, a B2C social app (like every other college kid), we heard so much about making sure you get the MVP (Minimum Viable Product) right. The MVP was the first major milestone we had to hit! I wish FirstRound would have published their article on MLPs Minimum Loveable Products back then; we would have thought through things differently.

As one of Entytle’s first employees, I’ve had a ringside seat to our journey to building a highly loved product — starting as an engineer, and then eventually as VP, Product. When we started, we wound up building an MVP first and then just built more MVP-like features without deepening our development on the core feature set. But let me confess, we didn’t really double down on creating something people love until mid-way through our journey. Now is a good time to also let you know that Entytle is a SaaS org operating in the manufacturing space — Industrial machinery manufacturing to be precise. Our customers (known as Industrial OEMs) are as Enterprise as they come. For those who haven’t heard of an Industrial OEM, these are the folks who manufacture machines that help other companies and organizations do various activities like heat/cool, inspect, make, package and otherwise perform critical functions in the enterprise. Ex. an Industrial OEM would be the one manufacturing the machine that helps bottle the Cola produced by another manufacturer. A typical Industrial OEM org is cautious, meticulous & far-sighted because their assets (equipment) lasts decades. At any time, they are supporting a 20-year-old machine while building the future product that’s expected to last for the next 20 years.

Now, you may wonder what changed “recently” that allowed us to make this shift from MVP to MLP given that our product has been around for three years. It essentially came down to two things -
1. completeness of vision and,
2. really listening to our users.

I believe that completeness of vision is in the hands of company founders — having a clear true north star can really drive strategic product development in the long term. We have phenomenal retention rates (upper 90s) that help us partner with our customers on our long-term roadmap. This has consistently helped us define the product direction & vision over the years.

As far as listening to users is concerned, that’s something a product manager can do (for that matter anyone who is close to the customers can do). We achieved the latter by setting up frequent user interviews (goal was once a week — not too frequent, yet not so far off that we lost track), sending out user surveys, and deep-diving into our Mixpanel data. The Superhuman product/market fit survey really helped us form an objective assessment of what we needed to focus on.

To prevent others from facing the same pitfalls we did, I thought it would be nice to share our thoughts around building loveable experiences (beyond just the product) for large enterprises (5,000+ sized organizations).

1. Implementation — Deliver value quickly
The implementation of your product could make or break the overall value delivered to the end-user. In addition to that, you also need to make sure your product can be implemented fast. No one wants to wait around for 1 year to see the result of their investment. At Entytle, our fastest implementation was less than 4 weeks at the height of the pandemic when manufacturing shut-down & customers were facing layoffs & furloughs. And that is what precisely helped expand to multiple new divisions within various customer organizations in the last few quarters. When the customer needs you, double down on the speed & quality to deliver value. It’s painful, it’s hard but if you are a smaller org that can pivot quickly, you’ve got it in the bag.

What really helps drive quick implementations are clear use-cases being defined early-on and having a dedicated project manager on the customer side. We would typically talk to this project manager at least once a week during the course of the implementation. Thanks to this clear implementation path, our customers are often generating a 2x ROI within one month of go-live. Goes without saying that in implementations without use-cases or clear project owners, strong ROI would remain elusive.

2. Training & Onboarding — Get users excited and get some easy wins
We’ve seen the wonders of a great onboarding experience with companies such as Superhuman. They’ve built a large fan base thanks to their unusual approach to training and onboarding every single user. Most enterprise tools tend to get complex, and most technology-savvy people (including me) tend to overestimate the ease-of-use of certain features. Something as simple as setting a password may still require explanation (yes, we see it all the time).

At Entytle, our onboarding actually starts 2 weeks before the actual training. We send teasers about our application and get people excited about the training. This makes the training more than just another calendar event, but something to look forward to. We also work with management and beta users beforehand to iron out any issues and ensure the customer has a process in place to incorporate our application in their daily or weekly workflows. This is critical in keeping a stable user base coming back to you each week.

During the actual training, we start off by setting clear expectations about what is in the product (from a data and feature perspective). Our demos are tailored to the use-cases provided during implementation, and we even use client nomenclature to feel relatable. People have a limited attention span, so we keep the first training light and focus on just a few core workflows or use-cases. We also show users how easy it is to achieve something quickly. Quick wins will really help with your user retention.

Pro-tip #1: During the training, we normally have one of our team members handling questions via the Teams/Zoom chat. This allows the trainer to remain uninterrupted, but allows users to get answers to questions quickly.

Pro-tip #2: You’d have to focus on building an onboarding plan for a week or two. Invest that time now, the dividends are huge, and repeatable & scalable to all your customers.

Finally, we use tools like WalkMe/Appcues/Intercom to give guided product tours once people log in to the application.

3. Proactive User Engagement & Support — Help users help themselves
While people may be excited to use your product after training, many of them will forget it exists within the first week. It’s not because they don’t like your product, but work and email get in the way, and then they simply forget. Even your most active users may not be using your product to its full potential. Therefore, it’s your job to constantly remind them how your product can help. We do this through proactive user-engagement and support.

To proactively engage users, we send across a series of email interactions immediately after first login, encouraging them to explore different areas of the product. We also remind users to log back in if they haven’t done so in 2 to 4 weeks. This works surprisingly well- more than 25% of users receiving these nudges log back in even after 4 weeks of inactivity. For the active users, we nudge them to adopt certain features based on how they are using the product (measured through Mixpanel and Intercom). For example, users might be using 5 filters when they can simply just search.

On the support side, we use Intercom heavily to let users chat with us directly from our application, as well as discover articles via our help center. It’s always amazing when the chatbot answers users automatically with suggestions from the help center (yes, this works!). For our larger clients, we have regular office hours where users can come and ask questions, or even share how they are seeing success with the product. In some cases, even after 3 months from go-live, we have a healthy number of users coming to office hours to ask questions or just learn from others. Ultimately, you can look at creating user communities (using tools like Tribe or Discord) where users across clients can help each other. We’re not there yet, but we hope to be soon.

Finally, we involve managers and leadership by sending them usage reports and learnings from other divisions. We have a QBR (quarterly business review) where we discuss how we can drive adoption company-wide and what they can do to support that.

Pro tip: Tamper your attrition outreach to take into account your customer’s business cycles. No two industries are the same & some tools are used every hour while some others will be used once a week. Find out where your product fits into the user’s workflows.

4. Configure and Customize — Let enterprises make it their product
Pretty much every large enterprise operates differently. They have different business processes, their own nomenclature, different KPIs etc. In order to create a product people love in the enterprise, your product must be configurable so that it can fit in with existing business processes and nomenclature.

For us, an “Opportunity” in one company, maybe a “Lead” in another. Some companies care about average revenue; some care about lifetime revenue. Therefore, we allow extensive customization of field labels and the order of display of each field (in tables and cards). Since we have Installed Base Management (Installed Base = Customer Base in Industrial OEM speak) capability, we also allow users to customize certain field types such as drop-down values, dates, numbers, etc.

Finally, one nice touch is that we allow companies to prominently display their logo on our product. Our customers really like this because it allows them to feel as if the product is their own. Some customers have even come up with their own name for the product (internally) and branding! Enterprise customers prefer pulling your product into their fold & make you part of their ecosystem — make it happen.

5. No Friction — Reduce the energy required by your product
Steven Sinofsky wrote a great article on “Frictionless” design choices and how important it is to reduce the overall energy required by an experience. This is incredibly important in the enterprise. Most users don’t know what that cute little icon is; be clear and put some text instead. Some key takeaways from Steven’s article that resonate with us at Entytle are :

  1. Decide on a default rather than options. For example, giving users an option to see a grid view or a list view is always nice, but start with a default, always. Also, think of this as ‘Convention over Configuration’.
  2. Create one path to a feature or task. Creating multiple paths not only causes user confusion but also leads to technical overhead when improving a feature.
  3. Stick with the changes you make. We’re always too tempted to redesign features, but we don’t account for the energy required to unlearn something old and then learn something new. Most features take months to get complete adoption, so avoid changing things too quickly if you’re not seeing enough usage. A feature we launched last October (2019) is only starting to gain traction after a few months despite weeks of nudging users to use it. We finally saw adoption when we had some early users get quick success with it and then let user-marketing run its course.

6. Simplicity — Make complex features feel easy
The consumerization of the enterprise has been a concept that’s been around for almost a decade now. And it still holds true today. Many large enterprises are working with legacy systems or extremely complex web applications that are very difficult for end-users to understand. You’ll be surprised how much you can differentiate by just having a better user experience and a cleaner interface. The biggest challenge here is that these large companies have very complex business processes and structures; therefore, the complexity of each feature you need to develop is quite high. Wrapping such features in an easy to use interface is no trivial task, but you will win the hearts of your users and buyers if you can. In many cases, the complexity of new features may be unknown initially, so always start with a simple base and iterate based on user feedback. This avoids over-engineering a possibly wrong solution.

At Entytle, we rolled out the ability to create saved customer target lists early on. This feature was as simple as it sounds, i.e., a saved list. We always thought users would treat these as leads and eventually convert them to opportunities in our system. The initial plan was to create a complete leads workflow to give users more functionality in the long term. Luckily we didn’t jump ahead too quickly because it turns out users wanted to create a list of opportunities right from the start (without having to save a customer list first). Imagine if we built all that complex functionality from the beginning; rolling it back would have been painful.

7. Dashboards — Feels like a “nice to have” but is a “must-have”
From what I’ve seen, most mature enterprise SaaS companies have the ability to show reports and dashboards. As a start-up, we often struggle to prioritize this use-case because it’s hard to perceive the value for end-users & time (read resources) is scarce to build a functionality that is usually seen as the bastion of decision-makers and not users. At Entytle, a large fraction of end-users actually look at our dashboards, and if we didn’t have it, it is likely that a ton of their deals may have fallen through. This is because the executives (who normally control the budget) spend most of their time in this section, even if it is just once a quarter. To sell to large enterprises, having some reporting and dashboarding capability is very crucial.

The good news is that you don’t need to build this capability from scratch; there are plenty of options to white-label/embed an existing solution like Looker, AWS Quicksight, Power BI. Have some pre-canned reports ready immediately after deployment, but be prepared to customize these reports/dashboards quite a bit. We normally schedule reports to be emailed to managers every week because they are more likely to look at it when they get an email. Finally, in order to support emailing and sharing of dashboards, make sure you can export reports to a PDF or even Excel if users want to do more analysis.

8. Alerts & Notifications — Send the right message at the right time
As mentioned earlier, many users tend to get busy with other tasks, so much so that they forget about your application. This tends to happen because they don’t see the need to do anything with your application at the given moment. The use of strong contextual alerts/notifications (email/desktop/mobile) can help keep your app relevant to the users.

Make sure every notification is contextual and personalized to the user receiving it. For us, it is important to notify users when a new lead is created in their territory or when someone shares a filter with them. To avoid notification overdose, we allow users to just subscribe to a daily digest instead. If you have a mobile app, then be sure to streamline notification delivery across platforms. Slack does a particularly good job of sending mobile notifications if the user is inactive on the desktop for a certain time limit.

9. Integration & Workflows — Move from a product to an ecosystem
Recently, I came through an amazing article by Ben Thompson, where he mentions the following:

Here’s the thing, though: Dropbox absolutely is better than One Drive. Google Apps are better at collaboration than Microsoft’s Office apps. Asana is better than Planner. And, to be very clear, Slack is massively better than Teams at chat. Using all of them together, though, well, it sucks: the user experience that matters for me is not any one app but all of them at once, and for the way I want to work, having everything organized in one single place is simply better (and that’s even with the normal spate of maddening Microsoft UI oddities!). In this, Teams is less a chat app than it is a file explorer for the cloud generally, and Stratechery LLC specifically.

As you can see from the above, the ability to have an integrated experience is incredibly important. You may have the best product in the world, but if it does not play nice with others, then users will quickly move away to a sub-par product that does. This is why having the ability to integrate with various systems is incredibly important. Over and above that, leveraging these integrations through out-of-the-box workflows will drive adoption faster. A workflow can be as simple as: when a user does A in your application, then do B in another application. Workflows and integrations go hand-in-hand if you want to deliver a unified experience in, and out of your application.

The tricky part with integrations in enterprises, however, is getting permission from IT. We have had the ability to integrate opportunities from our system to a CRM since day one, but in many cases, we have had to wait months before we can get our customers fully set up with an integration. The ability to have integrations and workflows constantly comes up during the sales process though, so make sure you can demo this capability.

Pro tip: Leverage companies like Zapier, Workato, Jittebit etc to help on the integration side and not build everything yourself.

10. Administration & Security — Strong security and administration are table stakes
Finally, the most important part of an enterprise product that everyone normally pays the least attention to — Administration. If you want to sell to large enterprises, they will all require some bare minimum administration and security capabilities before even considering you for an org-wide purchase.

Some critical administration capabilities needed are:

  1. Single Sign-On (SSO).
  2. Fine-grained roles and permissions for view/edit/download/delete on various core objects.
  3. Territories if your product is used by sales or service.
  4. Force logout and password reset capability.
  5. Enforced password requirements such as: 1 uppercase letter, 1 lowercase, 1 digit, 1 special character etc.
  6. Force change of passwords every 90 days.
  7. Multi-Factor Authentication.
  8. Usage reports and time spent in the application.

Important security measures to take:

  1. Install an SSL Certificate
  2. Encrypt data in transit and at rest
  3. Perform PEN tests regularly
  4. Maintain logs for all activity, but only store non-sensitive data in logs
  5. Limit production environment access
  6. Put strong firewalls in place and only open necessary ports
  7. Get a DPA in place for GDPR
  8. Get SOC2 certified. It is a painful and expensive process but goes a long way when dealing with enterprise IT.

To conclude, make sure you really listen to your customers and understand your market before building out any feature. The above points are themes I have seen across products from multiple verticals, but not all of them may hold true for the market you serve. Always ask yourself “why” each feature is so important for your customer. When we started Entytle, it was purely focused on building an AI engine for industrial OEMs to boost aftermarket revenue. We thought all the simple use-cases such as Installed Base management, accessing install base data, tracking expiring service contracts, etc., were solved for. We heard from multiple people that AI was the only way to differentiate. While it took some time, we realized that we know our customers best, and that led us to building an Installed Base Data Platform. Our AI engine evolved into a product-centered around retrieving, searching, and analyzing installed base data. You will be amazed at how many of our users just love the ability to find information quickly. Had we not listened to our customers, it is likely that our business would be a fraction of what it is today, and you would not be reading this post at all.