Fast Tracking your Engine for Growth
What we achieved through Bell Curve’s Growth Training Program
Last month, Astra completed Bell Curve’s four-week Growth Training Program. After coming back up for air – it really was an exhilarating and intense bootcamp-like experience – we’ve had a chance to gain enough altitude to understand the impact it had for the trajectory of our startup.
I’ll share details around why we decided to apply for the program, what it was like going through a growth marketing bootcamp, and the results we were able to achieve upon completing it.
Let’s start with what we achieved by taking the program: numbers, evidence, and finally an actual engine for growth.
By the end of the program last month, we saw a massive improvement across the board on our acquisition metrics. It has been exciting to see a step function change in growth rates, conversion rates, and even cumulative numbers. And that has meant crossing key milestones at a more rapid clip. A quick breakdown:
- 2x Improvement: Signed Up User Converts to Verified User
- 4x Improvement: Month over Month Growth Rate of Verified Users
- 4x Improvement: Monthly Active Users
- 1,000 Users
- 1,000 Automations Processed
- $100,000 Transfer Volume
Here’s what our growth looks like in pictures:
Because of the above numbers, and with our ads running with tracking in place, we also now have a statistically significant data set to draw insights and conclusions.
What’s our Cost of Acquisition? Are our value propositions resonating with potential users? How are our users behaving? And does that match our long term hypothesis?
We can now answer these questions with data.
For instance, we always envisioned that Astra would integrate with all bank accounts, but we assumed most users would be connecting their traditional big bank accounts.
In fact, we are seeing a much more frequent set of upstarts (aka “neobanks”) connected than expected – after Chase and Bank of America, Chime ranks as the third most connected institution, reinforcing our hypothesis about universality!
Lastly, and most importantly, we now have a proven engine for growth (even though we have only scratched the surface so far). We have a variety of levers we can adjust experiment more quickly, to grow faster, or convert better. Even though it was a lot of work to get the basics in place, Astra’s engine can work mostly in the background, it’s tunable, and even capable of further optimization.
Why would we spend time on this?
For Astra, 2018 was all about getting our MVP launched and learning from our users so we could iterate on our product.
By December, we saw signs of potential and stickiness, but we had one lagging metric: users.
Before the new year, we had passed 100 verified users on the platform and had 400 users overall. Month over month, we had a pretty consistent 27% growth rate, which was great, but we didn’t see it accelerating, nor did we have enough data to really understand the value potential users might see in Astra.
With the data we had, we knew:
- The product had to work without mistakes
- The friction to signing up was high, as it is for all banking apps
- If a user did verify her profile, she remained engaged. Bingo.
Breaking that down in terms of growth, we had to prioritize both quality and speed of growth. Our users have to share bank credentials and verify identities just to use the most valuable feature of the app, so we had a high bar to cross.
We had chipped away at growth experiments: a press release here, a co-marketing article with one of our great partners there; brute forcing by onboarding friends one by one this month; embedding ourselves in subreddits for that month.
It wasn’t systematic. It wasn’t methodical. We weren’t sure what to prioritize.
So while we felt we were well on our way to engaging users, the acquisition and conversion part was much more nebulous. Were our value props resonating? Did our onboarding flow have too much friction? Did people even care?
Enter Bell Curve
None of our core team had much marketing experience and this seemed like it would hit the key areas we would need to round out our growth equation. And we would be learning the content in a mentored program instead of outsourcing it. It would be a sizable investment for an early stage startup like us, but we decided that we’d be able to answer so many of our open questions while learning the growth skills we lacked. So we applied!
What the program is like
In short: prepare, get into a routine, and buckle your seat belt. 🏃♂️
The program has a trajectory but does not have a predefined track that all companies have to follow. You work with your mentor to align the specifics of the content they’ve produced to your needs, and then make progress each day through a feedback cycle.
We probably only touched 65% of the available material, yet we still were able to really achieve a lot across a number of fronts. It was definitely a reach for our small team, but the mentors, curriculum design, and infrastructure made it all work and result in a productive experience.
Here’s more on what that was like from a day-by-day and week-by-week perspective:
The Daily Cadence
For four weeks, Monday through Friday, the Bell Curve program had daily readings, objectives, and feedback touch points:
- The day before each training day:
- Read the material and do any associated research
- Start (and finish if possible) the deliverables for the next assignment
- The day of:
- Finish deliverables before the morning office hours on Slack
- Use morning office hours to ask any open questions
- Continue any open deliverables, check in with team members on associated items
- Use afternoon office hours to follow up on any open questions
- Start prep for the next day
- Catch up!
The program is described in three parts, but our experience really broke down across five weeks in practice:
Pro Tip: Do NOT underestimate how much time any one of these sections might take here and as much as you can, front load efforts so that all the revisions and reflection can be applied DURING the mentored curriculum. This is not a waterfall sequence – there will be additional items to tackle after each week.
Week 0: Prep (before program starts)
- Read Bell Curve’s Growth Playbook and use each section as a first pass exercise to what you will be doing
- Research and sign up for any additional tools you will need, like Hotjar and Heap 👌
- Align your team and get coordination tools/channels in place
Week 1: Strategy
- Create a customized growth strategy plan with your mentors:
- Which channels are a match?
- Who are your audience personas?
- What are your value props?
- Analyze the state of affairs:
- Your onboarding flow
- Your competitors
- Your website content
- Set up Conversion Tracking:
- Start this as early as you can to make sure it’s working
- More tracking data earlier is always good!
Week 2: Creative
- Redesign your Landing Page
- And start updating it ASAP
- We added a Onelink from Appsflyer and a “Text me the link” page for desktop users, which were both huge improvements for install conversion
- Design Ad Creative
- You’ll go through a few rounds of updates
- Below are some of our drafted versions
- Basic Ad Channel Setup
- Not supposed to start this week, but you’ll thank me for starting this early
- Develop against any changes to your Onboarding Flow
- Get this started ASAP as well
- For mobile apps like us, remember to incorporate the lag from App Store approval
Week 3: Acquisition
- Start running your first ads on your top priority channel
- Ours was Facebook and took most of the week to set up properly
- Tip: 3x your budget for the first few days to get more data earlier so you can make optimizations more quickly
- Start setting up second ad channel
- This was Google Ads for us – was waaay faster to set up than FB/IG.
- Do a first pass analysis on how your ads are performing
- Read up on additional channels and strategies from the curriculum – implement them if you have time
Week 4: Conversion
- Analyze your ad performance
- Dig into your own analytics for conversions
- Compare your acquisition funnel before and after the changes you made
- Optimize your ads
- Review each ad channel’s performance for each ad set and individual ad
- Kill your darlings if there is enough data and double down on what works
- Tip: We did an analysis and optimization on Tuesdays and Fridays, including updating our existing audience in FB, and we’ve seen a boost on new users after each tweak
- Dial back your budget
- If you juiced your spend in Week 03, don’t forget to dial this back
Now rinse and repeat!
For a company at our stage with a product launched but just hitting our stride, this program was invaluable in reinforcing our direction and boosting our confidence to take on our next big challenge.
If you’re a founder with similar questions, if you’re a startup trying to build your own engine for growth, and you’re ready for an impactful (and wild) ride, this program was a great use of our time.