Subscribe
Sign in
Can you review my GTM strategy for Chat GPT of Analytics? Steps being taken explained below 👇
I'm building Crunch (https://crunchit.ai). We're working towards building an LLM for analytics, starting with product event-based analytics. Work is on around integrations with different data warehouses. We also support an infinite canvas, which can help you plot multiple branches of graphs, sort of the way you would on a whimsical board. This enables you to answer any question about the particular data point. I've attached a demo to the working product here: https://shorturl.at/BGP48
We're doing the following for GTM
👉 Manually talking to mature but not enterprise data customers and getting them on a pilot. We're talking over 1,000 manually qualified customers purely through grunt work. The MVP is out :)
👉 Doing an extended waitlist through various content loops, e-books, webinars, tons of minuscule product videos and videos where I've my face etc.
👉 Working on publishable case studies with the beta adopters.
👉 A glimpse into a small offering via a drop-off detector going live on Product Hunt next week, but the original product with 100X more capabilities will be launched mid-October here.
👉 We've sneaked into over 20 renowned product communities, and we're not pitching anything. We're simply helping people with their queries and they end up asking what we do.
👉 Associations with product and data-influencers.
We're ruthless and saying no to most people unless we're dead sure they are the right pilot customer. We've no plans to run ads. We're getting fantastic feedback, but what else would you do if you were in my place?
The best comments get a waitlist spot even if you're early-stage ;) Let the feedback begin!
20
Replies
Your current strategy looks solid for penetrating the market, might I suggest speaking more often at relevant industry conferences and events as much as you can, if not already.
Also watched the demo, love the Magic Canvas approach! Is there any particular industry segment that seems most enthusiastic so far about Crunch? Are there ways pilot users have started utilizing Crunch that you didn't anticipate originally?
@yash_binani We have just started rolling the beta, so I'm looking out for surprises going ahead. We believe SAAS builders will like the solution the most because of the perfect balance that SAAS companies have in terms of their data to be analyzed and the kind of questions they usually ask and what we know our model to be good at.
Events are a strong yess but once we have established dominance. It may be not the right spend of our time and money at this stage.
Have you experimented with the approach where you look out for companies of a specific size and then reach out to their leadership team? Platforms like crunchbase are great to get you lists of companies doing >50M ARR along with direct contacts of their leaders.
@manas_tripathi That is an excellent point, and we have played around with a few of these sources. However, 50 M ARR is not set in stone because orgs with more ARR sometimes have no maturity and sometimes orgs with 50M ARR are too enterprise to start with. But I love your feedback. Thank you!
How much do you foresee this reducing the learning curve required to be able to use analytics effectively?
First off, congratulations on the progress you've made with Crunch! Building an LLM for analytics is no small feat, and your approach to GTM (Go-To-Market) seems well-thought-out.
I recently watched a video on GTM Strategy featuring Khadim Bhatti, Co Founder of Whatfix. ()
There's a lot of parallels to be drawn:
👉 Value Proposition: The video highlighted the importance of aligning a product's value proposition with hard business KPIs. Your efforts in manually qualifying customers and ensuring they are the right pilot customer align perfectly with this.
👉 Community Engagement: The speaker in the video emphasized the power of genuine engagement. Your strategy of entering into renowned product communities and genuinely helping out is a testament to this approach.
Few Questions that popped up:
👉Whimsical like Widgets is definitely an a-ha moment, but how long would it take for a customer to derive true value when using Crunch?
👉Feedback Mechanism: With MVP of Crunch released, how do you plan to capture feedback from early users, and what's your strategy for prioritizing and integrating this feedback into subsequent versions of the product?
Excited for Dropoff Detector! 🤩
This is for sure a very robust approach.
One question tho: How are you balancing customization vs out-of-the-box capabilities? Power users may want high configurability
@nayan_kulshreshtha We have a robust feedback evaluation mechanism at Crunch HQ. So we make sure any feedback is properly discussed and categorized and processed. However, in early days, more than often going for features like customization etc mean that we've not identified our target user base. Because a really good target user would find value without these highly custom features. Customization will be supported but not at the very beginning.
This is an interesting appraoch to take for GTM?
If am to be using mixpanel, how does your GTM for analytics product help me understand data better compared to an existing tool?
@balamurali I'll imagine you're asking what in our GTM heps customers get clarity on the value added. We have a playground. That playground works seamlessly on a demo data environment. At the same time, anyone hoping to try needs to invest 5 mins, and they would start seeing a significant value right off the bat.
Great Strategy.
What are the key criteria that you are considering while filtering out custumors for early access?
@sugandhraj_patel A matured data structure, A damn good reason not to stay with their existing analytics provider and on top, an appetite for payment because data processing never comes free. Also, the particular PoC works harmoniously if they have experience working with very early-stage products as a pilot partner.
The demo video was Interesting! Could you please provide further insights into the developer efforts required and the learning curve for using this product?
@mukund_chourey The development effort is insane, but thanks to LLMs, a small team today can also dare to start thinking of developing their own LLMs. Existing models are really good at certain things, and we're leveraging those strengths to tune our model. On top, the standard optimizations that go with reducing the response time when it comes to processing billions of data points or the standard effort behind ensuring that the best available advancements are deployed across the process of how we calculate the answer when a user types in the question in plain English stay as you would expect them to!
The learning curve would be minimal. We expect to be so good that as long as you can type a question in plain English, we should be able to give you an answer.
This is 💯 the way to go!
Love the magic canvas!
Warehouse integrations are definitely the key for smoother onboarding.
@ranjan_pai1 Thanks Ranjan. What else would you do if you were me? Or would you do fewer things? :)
Thats Intresting! How do you plan to sustain user engagement and retention as your product evolves and grows in complexity?
@okmegha By design, the product will not scale in complexity on the user's end. We want to do as much engineering as needed so that the other end of the funnel is strong enough to process any data. On the user's side, the UX will and should remain the same for the next whatever number of years. Ofcourse, there are advanced features that will come in but they will be unlocked based on proficiency so that a new user is not running into an unnecessarily high number of button.
DaoLens