Tasman

When you run performance marketing at scale it is essential to set up the best data and analytics infrastructure. Last year we worked with Tasman on building this and I am so excited to join Thomas in't Veld on stage at AntiCon to tell more about what we created together, the privacy nightmares we faced and the outcome of this project. See you in London!

Need data help, fast? Good news — we have a slot opening up in April for our fractional data team services! What is a fractional data team? Think of it as having access to a team of data experts, but only for the time you need them. Our Tasman team can help you with tasks like: - Data strategy - Centralising your data into a proper data warehouse - Data modelling and data analysis - Product Tracking and event analysis - Building dashboards & reports - Coaching and training internal data teams. Benefits of using a fractional data team: 1️⃣ Cost-effective: Get expert help without the overhead or cost of a full-time hire. 2️⃣ Scalable: Adjust your team size based on your project needs 3️⃣ Specialised skills: Access a team with a wide range of data expertise. Don't miss out on this opportunity! Reach out today and secure your slot now: 👉 https://lnkd.in/eCEXsHCe #fractionaldata #datateams #dataanalysis #datascience

Great article by Eric Thanenthiran and team discussing the best data stack set-ups if you are starting up or running a greenfield project in 2024. Go give it a read and don't forget the principles: separation of concerns, modularity, and don't-build-it-yourself!

🔊 DBT PACKAGE RELEASE 🔊 Read on to download our newest open source release — a dbt Labs package to model RevenueCat subscription data in Snowflake 🙌 Subscriptions are hard to get right, particularly on mobile. For any client where mobile subscriptions are a key part of their business, we have had to spent significant resources to make sure we modelled this data correctly. That's not as simple as it looks — subscription logic has a ton of edgecases and intricacies that can really skew any aggregate metrics like MRR or LifetimeValue. So, we built a universal logical model on top of the most popular source of mobile subscription data provided by our friends at RevenueCat. It's meant to be run on Snowflake to make optimal use of modern data structures, and it should be used as a logical data model that you can build any type of reporting structure on top of. Check it out here: 🔗 https://lnkd.in/eEjHcbDV

In 2023, Google began deprecating support for third-party cookies, and this trend will accelerate later in this year. While the loss of third-party cookies brings a number of new challenges (most notably around advertising performance measurements) it also brings massive opportunities (for instance for user privacy and better data quality). So, we thought we’d share our thoughts with you! If you are considering using a first-party data collection platform like RudderStack, then we've got you covered — Tom Shelley has bundled some of our best advice on how to maximize the quality and reliability of the data you collect. 🔗 https://lnkd.in/emDE5zxF #firstparty #cookies #webanalytics #GA4 #rudderstack

Subscription analytics is hard — particularly for mobile apps. Here’s why: 1️⃣ App stores provide limited info, making data extraction difficult. This is where tools like RevenueCat come into play. 2️⃣ Subscriptions aren’t static. They’re updated, plans change, exchange rates fluctuate, prices go up. Unlike one-off transactions, this data is always moving. 3️⃣ Metrics are not trivial 🙃 The latter point could use explaining — as the devil is in the details. Some examples: ‼ Active but canceling subscriptions (ie. under notice): Do they count towards MRR? ‼ Refunds: Should they be subtracted from MRR? Or are they a cost of sale? ‼ Monthly plans versus Weekly or Annual: Multiplying weekly revenue by four doesn’t equal monthly revenue. There are many different types of subscriptions as well to take into account: - Some are customer-level, others subscription-level. This affects e.g. how we view churn by product type. - Subscription types are diverse: freemium, tiered plans, licensed models. Each requires a different analytical approach. - App Store, Google Play, and Stripe handle taxes and commissions uniquely, affecting financials. - Most subscription apps sell globally, but have to charge VAT in different jurisdictions — complicating matters further. And yet, getting subscription data right is critical: 💰 It underpins all revenue reporting. ⚖ Incorrect revenue attribution can skew CAC, leading to inefficient marketing. It helps identify top customers for targeted marketing. 🛒 Pricing strategies and product initiatives rely on accurate subscription data. As with a lot of these domains, mastering subscription data is not just about tracking the right numbers. It’s also about aligning the business around shared definitions, and about making informed decisions that drive growth. Stay tuned for more information about subscriptions, culminating in us releasing a dbt Labs package facilitating data modelling!

At Tasman, the Snowflake data warehouse is one of our core data stack recommendations. Much has been written already about its abstraction of storage versus compute, its credit system, and the native support for nested data structures. But one of the most powerful features is its sophisticated access & permissions capability. All of you in larger data teams know exactly why that is such a big deal — different data consumers have different requirements, and you don't want to give full access to everyone all at once. But complex permissions have complex management, and the learning curve can be hard. So, our own Jovan Sakovic has written a fantastic explainer to Snowflake access control, kindly published by our friends at Select.Dev. Learn the difference between users, objects, roles and privileges — and finally understand what the RBAC approach means! 👉 🔗 https://lnkd.in/ggkmgYRf

So, you decided to set up your own modern data platform 🙌 . You've ran some POC's, made your infrastructure decisions, set them up, and have data flowing. Great! Your journey is just getting started :) Not only is it critical to think through the actual decisions that that stack needs to inform, you now also have a lot of data modelling to do. And most of the modern data stack discourse does not really help you with this. And yet this is the biggest pitfall for most nascent modern data teams — avoiding data model chaos. You need to design the models properly upfront, but what if you don't really have the time or resources to do that? What is the critical path approach here, and where are the trade-offs? At Øredev last November I gave our view at Tasman about this: how should startups approach data models? (Hint: it's not by scheduling a lot of ad-hoc queries...). See my talk below for some very concrete tips & tricks :) Data modelling for the win! https://lnkd.in/eC58fB6z

Event tracking governance has just changed in a major way and I am beyond excited! Read on to get started with our tracking plans for free. Good tracking governance starts with knowing what great looks like. So, we've partnered with our friends at Avo to have their new sandbox launch with tracking plans that we at Tasman use for typical ecommerce set-ups! That means you can now go to https://lnkd.in/emV_9RXR and immediately get started with a clear template. 🙌 👉 Here's what you'll get: a tracking plan that is highly representative of what our clients use in real production environments. The sandbox models a typical ecommerce website where users can add products to basket, create accounts, check out the basket, apply coupons, and get order delivery updates. There are also standard website interaction events (page views, link clicks, etc). We've talked in the past about the importance of having clear event prioritisation — so take note of the three Tiers we have defined. Tier 1 is critical for customers or revenue; Tier 2 is core product; and Tier 3 events describe UI interactions. It has been a very busy week at Avo — a full rebrand, amazing inspector features (speeding up event tracking diagnostics), fast tracking plan development, and much more to be announced at their ProductHunt livestream later today (see link in comments). Great work Tom Shelley with the rollout of the plan, and congrats Stefania Olafsdottir, Solvi Logason and team with a great launch week!

Event tracking governance has just changed in a major way and I am beyond excited! Read on to get started with our tracking plans for free. Good tracking governance starts with knowing what great looks like. So, we've partnered with our friends at Avo to have their new sandbox launch with tracking plans that we at Tasman use for typical ecommerce set-ups! That means you can now go to https://lnkd.in/emV_9RXR and immediately get started with a clear template. 🙌 👉 Here's what you'll get: a tracking plan that is highly representative of what our clients use in real production environments. The sandbox models a typical ecommerce website where users can add products to basket, create accounts, check out the basket, apply coupons, and get order delivery updates. There are also standard website interaction events (page views, link clicks, etc). We've talked in the past about the importance of having clear event prioritisation — so take note of the three Tiers we have defined. Tier 1 is critical for customers or revenue; Tier 2 is core product; and Tier 3 events describe UI interactions. It has been a very busy week at Avo — a full rebrand, amazing inspector features (speeding up event tracking diagnostics), fast tracking plan development, and much more to be announced at their ProductHunt livestream later today (see link in comments). Great work Tom Shelley with the rollout of the plan, and congrats Stefania Olafsdottir, Solvi Logason and team with a great launch week!