Andy Tyra

"We show them: 'Hey, you’ve got a shipment here that hasn’t been invoiced for 12 months,' or 'You’ve made a duplicate payment to a supplier,' and they go, 'Wow.'" Lars Reinkemeyer spoke to Alex Zaharov-Reutt about the process intelligence revolution and how Celonis is helping companies identify, frame and realize value. Watch the full interview here: https://celon.is/4gJF77E

3

Klarna has made a flurry of announcements about their new tools this year, but interestingly, they haven't provided any specific details on how they did it. If you replace 'AI' with 'awesome', 'crappy', or even 'God' in their announcements, the intent becomes visible. It seems like they're generating buzz to prepare for their IPO. If they've developed a cutting-edge CRM tool or innovative workforce solution, why not monetize it or offer it as open-source or freeware? That would create more excitement and marketing traction than these vague announcements. Are we to expect similar announcements about upcoming payroll software, bookkeeping software, and marketing automation software :-)

In case you missed it while prepping for Flag Day 🇺🇸 Happy State of Engineering Management Report Day happened this week! The annual thought leadership piece from Jellyfish has dropped. Hear from Engineering leaders, managers and frontline ICs .... we cover a range of topics that matter to great R&D leaders. Key Points: 1️⃣ Engineers are now strategic partners! 🙌 2️⃣ Burnout continues to be a challenge 😱 3️⃣ The talent shortage is spiking back up. 4️⃣ Many teams have embraced #AI 🧠 5️⃣ Others have legit concerns about AI ⁉ If #developerproductivity #genAI #engineeringexcellence are important to you .... have a read! (link in the Comments)

2

1 Comment

On this Tech Tuesday we're talking testing SDKs. Whether you're building SDKs or consuming SDKs someone else built, here are the topics to consider and ask your SDK vendor about.

12

Sometimes, you just need a quick way to visualize, right? 🔍 But every tool out there feels like a hassle—setting it up, learning it, or getting hit with that inevitable ‘subscribe now’ upsell. So I built my own in under 15 minutes. 💻 (Okay, maybe 25... 10 of that spent tweaking a two-column layout 😅), Talk about the 80/20 rule in action❗️ If you’re tired of overcomplicated tools, sometimes it’s just easier to build your own—in code, Slides, PowerPoint, or whatever works best for you. Check out this video and let me know what you think! 👇 #BuildTools

16

7 Comments

There is a tendency for many first time founders to overbuild an MVP. The reasons vary, but most often it stems from improper/lack of customer discovery and results in an improperly scoped build. The role of an MVP is not to capture the entire workflow of a potential user, rather it is to build a solution for a very specific white hot pain point that allows you to build trust with your customer and over time deploy more product and capture more wallet share. When a startup truly understands their target customer, and has isolated top priority problem areas, they are often able to zero in on a fairly narrow scope for an MVP. Oftentimes this solution can be fairly crude with a V1 UI/UX and not built to scale, but because it is meeting such a pressing need it generates robust usage from an initial cohort of users. On the other hand, when a startup fails to perform robust customer discovery, they almost always fall victim to the "feature fallacy" and overbuild initial products thinking that quantity of features will lead to adoption. It's the classic problem solving technique of throwing spaghetti at the wall until something sticks, but in early stage startups with very limited resources, this is often a path to failure. Founding teams need to balance the feature set of an MVP with their ability to ship product within the existing runway of the business - which in many cases is the sweat equity of the founding teams living off of savings or friends and family money they've raised. As such, the composition of an initial founding team is crucial, and having engineering talent on the founder team is most often necessary to bridge the gap. Founding teams lacking engineering talent and reliant on outside development contractors to stand up initial products put a startup in a suboptimal position to succeed. Investors are increasingly unwilling to fund first time founders to build out an initial product. As such, teams need to focus on a way to get a product in market and prove some degree of traction within their customer base before realistically targeting institutional capital. Raising money before deploying an initial product is possible, most notably by second time founders or early executives at high growth startups, but in many cases this subset already realizes the importance of customer discovery and getting an initial version of product in market, and optimizes for that before looking to capitalize the business. There are certain types of market opportunities where a narrow feature set is an unrealistic option for initial deployment - examples being an embedded payroll offering or a core infrastructure layer. In these cases, a founding team will need to raise capital to build an MVP. Second-time founders or founders with deep domain experience/networks who can demonstrate robust customer discovery proof points have a reasonable chance of raising capital, but most founders will struggle to launch these types of businesses.

111

40 Comments

🚀 2025 is shaping up to be a game-changer for product and engineering leaders. In areas like customer operations, automation has reached a tipping point. Now it’s about delivering that change into organisations. The best product and eng leaders will be the ones to drive that change. This buyer’s guide gives a mental model for picking the right AI solution for customer ops. 1️⃣ Start by recognizing the “iceberg” of customer operations 🧊 Many generative AI vendors focus on automating frontline customer interactions, however, it often only addresses about 25% of your total operations cost (tip of the iceberg). If you automate half of your frontline conversations, you’re only chipping away at a small fraction of your overall customer ops cost (e.g. 25% * 50% = 12.5%). The big opportunity lies beneath the surface: specialist support tasks, back-office processes, compliance checks, and complex judgment calls. When evaluating a solution think beyond the immediate customer-facing use case and consider how a vendor might help you tackle the more substantial bulk of your operational iceberg. 2️⃣ Insist on real-world tests over vendor-proclaimed metrics 📊 Almost every vendor claims their solution can deliver “up to 90% auto-resolution” and “up to 90% CSAT.” Those probably have 0 meaning for your specific use case. The only real proof lies in a live test. Run a controlled pilot, read at least 100 sample conversations end-to-end, and judge for yourself whether the tool actually solves issues. The difference between a great AI agent and a mediocre one is as stark as day and night. 3️⃣ Distinguish between co-pilots and autonomous agents 🤖 AI co-pilots, which assist human agents, sound great in theory but can be hard to measure in terms of ROI. Fully autonomous agents that can shoulder entire conversations can be easier to justify: their impact shows directly in improved auto resolution rates. Consider what setup best supports your operational goals and how you’ll prove the return on that investment. 4️⃣ Decide how much control and complexity you want to own 🏗️ Some platforms make you do the heavy lifting—building and tuning your agents from scratch, maintaining training data etc. On the flip side, they provide a great amount of flexibility in return. Others are more like a plug-and-play “AI brain,” where you only worry about giving the right business logic instructions (like you would with new human hires). Aim for a solution that aligns with the capabilities of your team and your willingness to manage complexity. Ideal solution will enable your ops SME to maintain and extend it independently. 5️⃣ Assess the vendor’s long-term alignment with your automation strategy Beyond current performance, is the vendor working in a direction that fits your long-term vision? Whatever the vendors offer today is just a fraction of what will be possible in a years time! What do you think? Any other useful considerations?

52

7 Comments

How do you keep engineering teams aligned with business goals while embracing innovation? Ada, the world’s leading customer service automation company, found the answer with Jellyfish. When pivoting to a generative-first strategy, Ada faced a key challenge: balancing innovation with supporting existing customers. They needed clarity on how their teams worked and invested their time. With Jellyfish, they achieved: 🚀 A 14% improvement in cycle times 🎯 Ranked in the top 97th percentile for “days with commits” 📈 Over 10% growth in technical investments “Jellyfish has become our go-to dashboard to understand how we’re operating as a team at a glance,” says William Mora, Head of Engineering at Ada. Now, Ada’s team is striking the perfect balance—focusing on reliability, experimentation, and investments that drive their business forward. Read the full story. Link in comments👇

26

5 Comments

What's the difference between functional and dysfunctional teams? Bradley Belford — Director of Product at SurveyMonkey is sharing his perspective on recent Growthmates episode 👇 "Shared representation and ownership are key to a functional team. When one function dictates the agenda, it leads to imbalances and missed opportunities. Encourage every team member to contribute to the backlog. Regularly pause to ask, 'What are we missing?' and welcome diverse inputs. This approach ensures that all perspectives are considered, driving more valuable and user-focused outcomes." Find more insights beyond this in a full episode: 🔹 SurveyMonkey Growth Evolution over the last 8 years 🔹 How they transitioned to Product-Led Growth 🔹 What's the role of Design and Research with valuable insights from Erica Weiss Tjader … and more Listen now on Apple, Spotify and YouTube (links in comments) 👇 #plg #growth #collaboration #teamwork

37

4 Comments

This is a fabulous, hands-on, practical review of Helen Bird's first year in Product Operations. One thing I dislike these days is this constant needs for ultra-high level, ultra-generic articles and descriptions of jobs. What I really like here is the detail, the in-the-weeds problems solved. Because we are all professional and practical enough to take what she have done and apply it to our own situations. This is a great read for all aspiring or early-stage Product Ops professionals. Well done Helen.

8

1 Comment

SaaS pricing models are... kind of a mess right now, aren’t they? The traditional per-seat approach often gates access unnecessarily and fails to reflect the diverse ways teams use these tools. Instead, pricing could focus on roles or outputs, charging more for creators or power users while enabling broader organizational access. And when you layer in GenAI, things get even more complicated. Fixed-price GenAI tools don’t align with the unbounded costs tied to their underlying LLMs (just look at OpenAI’s recent $200/month tier). For SaaS companies, this creates real challenges around balancing customer value with profitability. As the EY article highlights, navigating this complexity requires SaaS companies to shift from being just vendors to becoming strategic partners. At Higharc, this collaborative mindset shapes how we deliver value. When companies focus on aligning pricing with usage and fostering stronger customer relationships, the opportunities for growth are immense. #SaaS #GenAI #PricingModels #Collaboration

5

4 Comments

Three lessons from building and launching a P2P payments app without doing the proper customer validation first (despite knowing better): 1) Building something you need is a great starting point - but not enough to validate an idea. As someone drawn to building solutions, I’m an early tech adopter. This led me to overestimate consumers’ willingness to onboard, especially with a P2P payments app that requires KYC/KYB compliance and relies on a network effect to be useful. 2) Manual solutions are often best to start with It’s good business to be in business - transitioning from ideating to operating as early as possible is the BEST way to learn effectively. While the move fast and break things mindset has attracted well-deserved skepticism, getting stuck in a multi-year build trap distracted me from critical real-world validation. 3) Consumer appetite shifts quickly In my case, the proliferation of free and instant P2P payment options like Zelle added additional pricing pressure that I couldn’t compete with. This disincentivized the consumer event/payment collection market, turning my bottom-up go-to-market strategy upside down. Pitch (https://www.pitch.fun) is currently on the back burner as I’ve begun building custom solutions in various industries to sharpen my skills, solve specific problems, gain a hands-on understanding of operations, pain points, and workflows, and build a network of experts to lean on as I keep my eyes peeled for high-impact opportunities.

11

Bring your Planning use case and challenge Martin Kostic and the Inphinity ∞ Team at Qlik Connect next week to see how they can help you!

8

Start-up rules - Half as long - When your development team gives you a ship date, push back and ask them to get it done in half as much time. Pushing back forces your team to reconsider their all their assumptions and often gets them to deliver better products, faster. I grew up the son of an english teacher. My mom was nice enough to proof most of may papers. Her frequent input about my papers was "half as many words"..... If you have seen the movie A river runs through it, the main character's dad often gives his son the same advice about his son's writing... half as many words... Shipping new products is a bit like writing an english paper, often the development process is too long. Between feature "creep" (adding on too many features), conservatism (I need to make sure I pad my time estimate to make sure I hit the ship date) and too many cooks in the kitchen, the project end date slips. At Tripadvisor, my partner Steve Kaufer taught me a novel project management trick. When someone gave him a delivery date, he would almost always ask for the product to be delivered in half as much time. The process of asking for the product in half as much time challenged the team to reconsider all their assumptions - reduce the feature set, accelerate the ship date, accept some flaws in the initial "beta" product... maybe the first version doesn't need to be perfect..... In managing your development teams, don't just accept the delivery date given. Push the team to do it in "half as much time". Doing so, will often uncover a tighter process and feature set.

38

2 Comments

Klarna firing their SaaS providers is just the beginning. Executives have been weighing the build v. buy decision for software and IT infrastructure for years. Now, no-code and low-code, better AI and accountability mechanisms like economic bonding and restaking make it possible to build something better than what one can buy. Interested to see where this goes. What do you think of Klarna ripping out Salesforce and Workday in favor of internal tools? #ITinfrastructure #techstack #buildvbuy #saas

10

3 Comments

Pain point doesn’t mean leverage point. You need to look at both in platforms and infrastructure. As you stack leverage points (by building good abstractions) the ability to remove pain points quickly compounds. Balance the investments in pain vs leverage fixes. If you don’t you risk tech debt with quick pain point fixes OR losing faith from users by taking too long

27

1 Comment

Highly recommend this course for anyone who wants to wrangle, ahem I mean develop, people brilliantly!

7