Case study: Developing a recognition platform for a technology company.
A case study on building a successful HR Tech Product to promote employee recognition.
TL;DR Summary
The existing recognition platform had some pain points around integration with other systems and the latency of the rewards. The timing was right for a replacement.
The new solution leverages existing tools and channels, more immediate rewards, is less bloated, and since it’s based on a free feature, allowed to repurpose the license fee cost to support a bigger employee rewards’ budget.
The product was quickly adopted and participation has sharply increased in all the relevant metrics (more total nominations, more people actively nominating and being nominated, more inter-team nominations, time-to-value, etc).
Some Context
About 15 years ago, [Company] helped build an application for employee recognition for a local startup. [Company] became a client so us to “eat our own dog food”, and it was hosted on [Company]’s servers for free in exchange for the usage fees, so it was a wash. It worked relatively well like that for several years.
Some of the benefits of the application:
customizable to fit [Company]’s values and behaviors
rewards’ redemption was self-service so it only needed to periodic balance refill
However, after a while -as most hosting moved into the cloud- [Company] wound down its hosting business and this app moved into the cloud, so instead of being a wash, it became an annual cost.
The problem
Over the years we had observed several pain points in how we used the application and when the vendor approached us to upgrade the platform to their new version -because they had been acquired by a larger company that was sunsetting the legacy platform- we saw this as an opportunity to upgrade the experience that our users were getting out of it. We explored the newly proposed solution we were being grandfathered into, and while it offered a moderate improvement, it still suffered from some of the same problems.
We interviewed several stakeholders (Executive team, HR team, and Employees) to determine the key pain points to solve.
They were determined to be:
1- lack of integration with current systems: users had to log in specifically to the employee recognition tool (which had a separate, non-authenticated login) to use it, or to view all other recognition activity.
2- low adoption/user activity: as a consequence of the above
3- limited impact: most nominations (recognition instances) happened within teams, and there was no highlight on the collaboration across teams.
4- gaming the system: since both nominator and nominee were getting points we detected some presumable instances of“quid pro quo” nominations.
5- latency between the action being rewarded and the prize awarding: users received a small number of points, which had to be accumulated to be redeemed at a later date. This diluted the recognition impact.
6- unnecessarily complex reporting: reports were somewhat painful to extract and required further processing on other tools.
7- the cost of operation: it had a fixed annual cost of operation, on top of the budget dedicated to rewards (the budget for rewards constituted less than 20% of the total cost, the rest went into license fees).
Exploring approaches
We identified that problems #2 and #3 were dependent on #1, so that became a high priority for our search of a replacement: whatever the solution, it needed to live within one of the established tools people used every day. In our case, the best candidate was MS Teams. So we started to explore and demo 3 different solutions that integrated with (or lived within) MS Teams.
We observed that problems #4 and #5 were related: the points were poor proxies for value and provided a very delayed gratification. At the same time, while rewarding both actors with points aimed to encourage system activity, it also generated distortions on how the tool was used, that did not align with the engagement goals for the platform.
We saw problems #6 and #7 as opportunities: we only needed some very specific reports, so we could easily scale down our needs; and if we could find a cheaper solution, we could put more budget into the rewards.
After reviewing and doing pilot experiments with 3 different MS Teams-based solutions, we noticed some common patterns:
The solutions struggled to work well with MS Teams, they were unstable, probably due to the fact that the marketplace and ecosystem of MS Teams apps was quite new at the time
Most of them relied on points, and they tried to force-feed their partnerships with rewards vendors.
The Solution Process
The search, nevertheless, was fruitful because in the process we came across a free, rather obscure feature of MS Teams: Praise.
Praise allowed any individual in the organization to create structured MS Teams postings on a specific channel, which included a title, an icon (customizable from a list of available ones), and a text blurb, tagging the corporate value being recognized, and the people being recognized. That covered most of our needs when it came to recognition. The remaining problems to solve were Reports and Rewards.
We implemented Praise through an MVP, which we released to a small group of volunteers, including some that were power users of the old platform. The objective was two-fold:
Gather feedback on ease of use and concerns from the team.
Gather enough utilization data to be able to explore the generation of reports.
A 30-day pilot was conducted, and the initial results were satisfactory. The team found the product easier to use, and we managed to gather some dummy data to build initial reports on.
With one developer assigned to the initiative, backend reports were created to identify:
Inter-team vs intra-team nominations, source and destination departments of nominations
Popularity of nominations (based on social media activity on MS Teams)
Top nominators, top nominees
# of total nominations within a time period
# of nominations attached to each of our corporate values
Historical trends
We defined 2 key metrics for success:
Total number of nominations per month (as a proxy for individual engagement and participation)
Inter-team vs Intra-team nomination (as a proxy for team engagement and collaboration)
We established a timeline for launching the new product, overlapping with a sunsetting period for the old platform (2 months). During the sunsetting period, users were encouraged to start using the new platform and to use up the points balance they had in the old platform. Unused points were to be converted to dollars and donated to a charity (which greatly reduced the FOMO some employees had expressed at some point during the initial interviews).
Prior to -and during- the implementation, we identified some additional potential risks/challenges and the ways to address them:
-Team resistance to change (perception that they were losing something with the change). We addressed this with a thorough communication plan that explained the reason for the change, how the process would be, and what were everyone’s options (redeem up to a certain date, or leave the points and donate them to a charity).
-Logistics and variety of the rewards (without self-service, somebody had to manage these manually). We addressed this by:
creating a clear process and accountabilities within the HR department to manage the rewards
reducing the workload by limiting the awards to 5 nominations per month (and 5 Quarterly super winners), who would get bigger (fixed $ amount) instant prizes, instead of ALL nominations getting a small number of points to be redeemed at a later date.
Creating a survey for the winners, with a limited list of reward options (mostly gift cards, since it’s a globally distributed team), which includes one option to “Try a new reward”, which enrolls those who select it in an experiment with a new reward type. Since they are voluntarily enrolled, this reduces the risk of dissatisfaction in the experiment.
Limiting the number of nominees per praise to 3, so that the (fixed) amount of the reward does not become too low for each individual. This had to be done manually, as Praise does not allow such limitations for the time beint. But it was justified for 2 reasons:
Nominations of large groups usually fail to represent the individual value. More individual recognition is more powerful.
People are still free to nominate more than 3 if they want; however that would render the nomination ineligible for rewards. The offered workaround was to split large nominations into several small ones to adequately depict the value created by the individuals being recognized.
-Confidence in the rewards system (perception that the rewards were being awarded fairly). We addressed this by assembling a committee of volunteers that review and choose the winning nominations. That way, the team has participation (through representatives) in what nominations get selected.
-Lack of dedicated development resources for internal projects resulted in having to work with 3 different developers over the course of this 5-month project. We resolved this by having knowledge transfer meetings with each new developer
The Results
The whole process of replacing a 15-year legacy platform took 6 months, from idea to launch. The launch and implementation were painless, and adoption increased 200% in the first month (later on, it stabilized around 150% of what it was prior to the launch).
Additionally, it was observed that The Time To Value had decreased significantly: new employees organically started utilizing the feature early on (during the first month of employment, sometimes during the first week), which in the past required heavy-handed onboarding and didn’t have much participation until 3–6 months of employment.
Some items backlogged for future improvement:
1- Delivering automated newsletters and reports: reducing the human effort to generate the weekly newsletter and monthly reports (requires further investment in backend development to be approved).
2- Enabling more variety in the rewards for a remote team (right now is limited to Amazon and Apple gift cards) without increasing the human effort or completely de-personalizing the interaction.
3- More and better guidance for users (aided nomination process, nudges for participation, limiting the # of nominees per praise, etc.)