Being Data-Driven Doesn’t Have To Be Hard: Adopt a Powerful Tool

Josh Dawson

Head of Data at TouchNote

people reacted
6 -minute Read,

Posted on April 28, 2022

Multiple teams across this 14-year-old scale-up have gained the confidence to analyze data and answer their own questions.

Touchnote

It doesn’t have to be hard to get product-level insights. You don’t have to have a huge budget for computing power or employ a large team of business intelligence (BI) professionals to write scripts. Instead, you can use someone else’s scale to accelerate your business, just like we did.

Everyone wants to become more data-driven— the buzz phrase de jour — but many businesses don’t have the resources to do this on their own, or they don’t support the heavy lifting it would take to make a complete transformation. I came to TouchNote with a mandate to turn that buzz phrase into a reality.

The Growth of Our Data Team

Over the past 14 years, TouchNote has built an easy-to-use app to create and send custom-made cards. When I joined the company, there was only one analyst trying to answer everything related to data. I wanted to teach the business to become more self-sufficient with data, but everyone had a ton on their plates. Getting everyone to learn a new skill set around data analysis to become completely self-serve was too big an ask. Instead, I set out to grow our data team and simultaneously improve people’s capability to answer their own questions.

 You don’t have to have a huge budget or employ a large team to become data-driven. Instead, you can use someone else’s scale to accelerate your business.

The first step was to bring on more data professionals. When I joined the company in 2020, the team consisted of one data analyst and myself. Today, we are a team of seven across India and the UK with expertise in data science, predictive analytics, modeling, and software engineering. Our insights team started out answering people’s questions and building dashboards to create stories that explained the data we saw. Meanwhile, our engineering team underpinned everything and ensured data came from and arrived at the right places.

We underwent a few big projects, including redefining our business logic and bringing everything together in Looker, our BI tool. Once we had that under our belts, we wanted to keep the momentum going and turned to events. Those, however, were a big mess.

A Familiar Platform to Help with Time-Consuming and Labour Intensive Tasks

Young startups only have a little bit of data to handle, but given our 14-year history at TouchNote, we have years and years of data at our disposal. But, that means event tables are incredibly hard to write and manage.

The problem with traditional event tables is they are usually enormous. The experience of going online and searching for a dog bed, to purchasing that item might log 60 events—and that’s just for a single purchase. Multiply that by however many thousands of products a company sells in a year, and it means that data quickly exceeds the size of the business. At that point, unless you want to spend a load of money on computing power, you end up having a queue of queries, and any question someone asks goes to the end of the line. Every question leads to more questions, requiring further portioning of the table.

You can perform a one-time analysis on that funnel, but it takes two or three days to execute the query. There’s no real-time visibility into the business, and on top of that, it’s a lot of engineering work. Another option is to create smaller event tables, but that’s not as powerful—and our engineers didn’t have enough time to do that.

Instead, I turned to a platform I’d used at other companies: Amplitude. I knew the value of its power and simplicity. It’s a point-and-click tool, so users don’t have to learn to write SQL to glean insights. Once I knew what we needed, the next step was to figure out a way to leverage the capabilities within Amplitude to serve various teams.

Teaching Others How to Fish

I believe getting data into everyone’s hands will push them forward in their data journey, but it’s not always so simple. Take two different golfers: an amateur who plays a casual round on the weekend and spends most of their time at the clubhouse, and a pro who seeks out competition and spends hours working on their swing. Even when they use the same tools, the competitive player will score better every time.

It’s the same with data professionals. We spend years learning how to obtain and use data, so we think about it differently from the casual user. But we can still show an amateur some tricks. As a data professional, you have to be open to teaching others rather than doing everything yourself. Otherwise, you’ll never have any free time to do anything aside from answering people’s questions.

Data professionals have to be open to teaching others. Otherwise, they’ll never have time to do anything but answer simple queries.

We initially ran managed trial with our acquisition and activation squad, one of our product teams. They already had some experience with data, and we knew they wanted insights into their funnels. After we officially adopted the platform, we grew adoption by releasing events focused on specific teams, such as growth and finance. Once we got to the point where various user groups knew how to use Amplitude for their particular needs, we launched a weekly clinic for anyone to drop in and ask questions.

We found that if you teach teams well from the start, managing analytics quickly becomes everyone’s responsibility rather than just that of the data team. Thanks to our training, usage throughout TouchNote has grown to 80–85%.

Our Teams See the Impact of Their Work

We quickly got to a point where we were data-driven in both definition and practice. Everybody from developers to QA teams use Amplitude because everyone knows how to obtain data easily and quickly. All of our attribution is in Amplitude, so our growth team can split users by channel and understand the behaviour of those users. For example, they can discover whether certain channels lead to hyper-engaged users or whether users from certain channels aren’t as active as we’d like them to be.

Even our monetisation team is in on the act. The biggest event we launched in 2021 was a gifting program. We decided to create a dashboard to track gift buying in real time, which created a buzz around the program—and the dashboard. From there, curiosity took over, and people started exploring the platform independently.

Amplitude has also helped us ramp up our A/B testing, which has been massive. Now we can see the effects of moving a panel to the bottom or scrolling right versus left—all those minor changes we make all the time. Running similar experiments without A/B testing used to be extremely slow. Even the process of submitting a query to our data team used to be a massive conversation that took place over several weeks. Now, a novice can click around in Amplitude and answer their questions within seconds. We’ve found and fixed a lot of bugs this way too. The speed at which teams can move in Amplitude is one of the drivers of adoption because it helps our teams improve and see the impact of changes in near real-time.

Features That Increase Visibility

TouchNote isn’t rocket science: we sell greeting cards and gifts. Even though we’ve been building this app for 14 years, we regularly make small changes. We need to make our card flow process as straightforward as possible: a user starts a card, creates a card, then sends it. When looking at the card-creation flow in Amplitude, we saw that people started, designed, and completed a project, but they looped back to the start because they hadn’t understood the differences in our offerings—in this case, the difference between a card and a postcard. We realised we needed text explaining each card option earlier in the process to push people through the funnel.

Funnel analysis charts are among the most widely-used features within Amplitude at TouchNote.These charts give us the chance to look over our users’ shoulders and see how they use our app. Our users have high intent—you probably don’t download an app to send cards if you don’t want to send cards—but with funnels, we can understand their likes and dislikes in real time.

Having teams answer their own questions saves time, and we have increased our signup-to-paying-user conversion metric by 66%.

A key metric for us is the conversion from sign-up to paying user. Many of our business strategies are born out of that conversion, so we have to understand the user behaviour that impacts that metric. From our experience, influencing user behaviour is often less about introducing something new than identifying and fixing pain points. There’s a lot to be said for not pissing off your users! We use Amplitude to identify and improve these pain points quickly, and as a result, we have increased our signup-to-paying-user conversion metric by 66%.

Our Business Is Better Because We Found the Right Tool

Our high adoption rate is a significant win, but I’m also impressed by how people have upskilled so quickly. I’ve watched our non-data professionals who were almost afraid of crunching numbers become more comfortable asking questions of data and working with it.

Fantastic work is happening at a rapid pace at all levels of the organisation, from product managers to developers. We receive outstanding support from our C-level and the team at Amplitude, and we draw on those resources whenever we need help. Everyone outside the data team now drives Amplitude usage, and I’ve learned that anyone can drill down and analyse data as long as they have the right tools and understanding.

Anyone can drill down and analyse data as long as they have the right tools and understanding. 

Amplitude has given everyone at TouchNote the confidence to analyse data and answer questions. Every day, teams learn more of what they can do and what they need to do to improve the app, which keeps everyone excited, engaged, and motivated to keep moving forward.


 

North Star Playbook Ad CTA

Josh Dawson

Josh Dawson is the Head of Data at TouchNote. Josh leverages statistics, ML, experimentation and a strength for visualization in order to provide concise business recommendations.

More from Josh

Inside Amplitude