December 19, 2023
Effectuation and a CRM for Strategies
Daniel Sternegard
Director of Product, Planhat
Effectuation
Being part of building a company reminds me a lot about a concept I (randomly) ended up writing about in my bachelor thesis: effectuation. It’s perhaps unsurprising that I keep getting reminded, since the theory specifically talks about decision-making in entrepreneurial contexts. Either way, it resonated, stuck, and comes back every now and then.
Effectuation states that when making decisions in high-uncertainty contexts, decision-making turns from prediction to action and control. If context is changing fast enough, then don’t spend time trying to analyse what will happen in 5 years, because you can’t predict that - spend time on taking the best available action in front of you. Pick up the shovel and start shovelling. But importantly, with a loose long-term vision in mind. You’re not floating free without purpose, you’re just focused on taking the next best action given a defined direction.
Continuously refining the vision
There are a few very interesting points in the process of building something when the higher-level vision is re-articulated, updated.
It seems likely that these moments are hard to predict, sometimes coming with a flashing insight and other times as a gradual re-definition. These moments often don’t mean a complete reversal vs. the previous vision; rather, they represent a refinement of the existing thesis. In the accounting world, this is called “hardening” - objectives are concretised as a result of learning.
To me, such a moment is deeply personal. Each person has their own long-term vision of where the world is heading, even in an organisation that is highly aligned. Narratives and dreams are always personal, weaving in each person’s unique motivations and understanding of the world. All of this is true in every aspect of life: from personal dreams and decisions, to organisations, to society. The constant is that in all domains, many of us have long-term visions on where we should go, and that vision is continuously refined.
In the case of Planhat, I have previously written about A Data Platform for Action (with Kaveh Rostampor) and about what a Customer Platform is. Each of these were true in the moment and to a large extent still are. But after another year of learning, spending days and nights in the fight, now again I feel is the time to refine and re-articulate the vision. This is by no means the only vision of the future at Planhat, and even though it’s highly aligned with the direction others believe in, it’s also deeply personal in how I choose to express it. That diversity is healthy and welcome.
But hopefully this article can serve as a clear summary to myself, to current and new team members, and to current and new customers, on where Planaht is going.
Big trends
If a trend is big enough, it stays true for a long time.
This is true for the growth of software and digitalisation. For the growth in volume of data generated every single day. And for the growth of businesses moving to more long-term and relationship-focused models; where you are less evaluated on rapid short-term growth than on something sustainable, that can attract and retain and grow customers.
The latter trends serve as the primary need for businesses: how do we become effective and sustainable over the long-term? How do we sell to the right customers, in the right way? How do we then serve, retain and grow them in scalable way? If you can do this, you have a right to exist.
The former trends serve as the input for solving that need: how can you leverage data - especially time-series data - and software to become even more effective?
CRM for Strategies
It seems reasonable that at the intersection of those trends lies something curious.
Software grounded in data that can help you understand who to sell to, how to sell to them, how to then best onboard, support and service them. Crucially, software which houses not only the analysis of data, but also the execution of strategies; when you have both, that’s where virtuous cycles begin.
Such a platform includes a few core parts. You need to 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes.
This is what a CRM should be, and it is - perhaps surprisingly - the most strategic and exciting tool that you have ever seen in business. Building a CRM is building the platform that all the world’s companies will use to deploy data-driven strategies that make them smarter and more effective. How will decarbonisation platforms commercialise? How will green steel be sold? How will OpenAI onboard enterprise customers?
This is actually what our go-to-market teams deal with every day, speaking with leadership at customers on what their biggest commercial priorities are, and help them operationalise those strategies. Do you plan to expand your team next year? Do you plan to move part of your sales organisation to upsell and cross-sell existing accounts? Are you launching a new product? All of the resulting strategies end up in a CRM, because it’s where you can truly operationalise them.
The fascinating part is that given where the trends are, this wasn’t possible to build some years ago, but is now. Data is maturing, and with falling compute costs, processing vast amounts of time-series data is now commercially viable. The speed at which we can build frontend applications is growing by the minute. The growth of Work OS tools (Notion, Monday, Clickup, et al) has brought consumer experiences to enterprise tooling. And at the present moment is - pretentious as it sounds - a generational opportunity to build something incredibly leveraged and impactful to modern business, combining these pieces in thoughtful and creative ways.
To me, this is what Planhat is building: a place where you 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes. Note that this is a natural cycle where your outcomes continuously become your initial customer data input, and your strategies become more optimised as a result of that learning.
This is part data platform, part workflow platform, with the power of the former and intuitive experience of the latter.
Workflow Economics
I am weirdly but proudly passionate about what the market calls Workflows. Workflows are, well, the various flows of work each business runs. You might as well call them processes, but workflows probably sounds a bit more hip.
I am passionate about it because workflows represent the action you take; they represents the forward momentum of your business, and - when perfectly executed - they represent the best version of what you are trying to do. How do you onboard a governmental agency to your security solution? How do you run the design cycle when doing a proof of concept in the electronics hardware industry? Which segment should we go after first when commercialising our new line of industrial robots? What are we doing?
Your workflows, or actions, are what generate outcomes, which is what you care about. Workflows are then combined into overarching strategies that you deploy to reach objectives.
Workflow Economics is a concept that came to mind, which fits very neatly into this cycle. If every business is about running workflows to meet some objectives, doesn’t it make sense to measure which workflows seem to be meeting those objectives and which are not? If you have two ways of onboarding a customer - either self-serve or through a high-touch onboarding - doesn’t it seem pretty critical for you to understand which is more effective at driving adoption, or year 1 renewal (depending on what your objective is)? Or, maybe, if you are meeting all of your objectives (eg, adoption) but you still churn the customer - doesn’t it seem reasonable to understand why adoption is not driving revenue? Similarly, you want to understand the cost side of things, like how much time are we spending with each customer per week?
Workflow Economics is a lens through which you understand the effectiveness and returns of the various processes you run. The processes you run is ultimately who you are as an organisation. So Workflow Economics is what allows you to dissect which parts of your strategy and operations are being successful, and which are not.
Needless to say, Workflow Economics is a core concept to what we are building at Planhat. Planhat is built for you to set objectives, define and execute strategies, and measure outcomes. Workflow Economics is the engine for optimising this cycle.
AI AI AI
Everything today is about AI.
It’s an area that I, like most others, am very curious about - but by far no expert in.
But looking at the intersection of the AI fundamental rules that I do know, and how I think about a CRM for Strategies, you could state some interesting hypotheses.
One such fundamental is that AI is only as good as the data/context it gets. Try prompting ChatGPT, and now try prompting it with 3x the context on the problem, desired outcome, and parameters to consider (or pre-train it on a more relevant dataset). The answer gets significantly better.
It logically follows that the success of AI applications will largely depend on how well you can provide queries with relevant context. By context I mean data on what you are trying to solve, what your desired outcomes are, what has worked well in the past, and what some of the parameters to consider are.
For example, consider prompting some LLM with “what is the next best action to take after this call” and provide it with a summary of a call transcript. Then prompt it with the call transcript, and all the ticket data for the same customer for the last 30 days. Now add all data on how they are using your product. Now add relevant external data, from outside your organisation. And, crucially, now add data on what next best actions has worked historically in similar contexts.
Again, don’t take my word for this - I would love to hear counter-arguments - but it seems logical that, at least up to a certain threshold, accuracy scales with the volume and quality of contextual data.
The implication of this is what? First, I don’t know where the current AI progress will plateau or lead us. Second, knowing that, I bet that a system which is built to consolidate and synthesise all customer data, which contains data on your objectives, and the outcomes of previously executed activities, will be helpful in leveraging AI for benefits.
Architecture
Something which is definitely not my vision, but something I have simply wholeheartedly adopted from our technical founders, is how to architecturally think about building such a software as described above.
It needs to be generic and horisontal. Build modules or components that represent real world concepts, that can be assembled to solve a high variety of problems. And make these pieces interconnected - that is the critical part. Data feeds into segmentation feeds into workflows feeds into analytics and feeds back into another segmentation which feeds into… Close the virtuous cycle of data and action, as measured by workflow economics, and grow.
This is what you see when peeking into Planhat’s technology. An architecture built on data, workflow, automation, and interface platforms. All connected, with a unifying permission and security layer.
The architecture sets the basis for the long-term ability to innovate and expand the product. I am genuinely happy to be working with a team of engineers who have from day 1 built something that I can now be part of innovating on top of.
The fascinating part about effectuation theory is that while it emphasises value in having a clear vision, you’re never any better than what you can cobble together each day. To move an increment forward, knowing vaguely where you want to end up, but with a laser focus on the action. There’s an entirely separate article to be written about culture at Planhat, but this mindset is definitely deeply embedded - don’t get lost in theory, pick up the shovel, and let’s move.
But some key themes have clearly emerged in my thinking, over the past years of building a CRM, that provide overall guidance.
First, effectuation is a real thing. Continuous refinement of vision is good.
Second, building a CRM is the most exciting thing there is, if you think it’s exciting to think about the evolution and operation of modern businesses. How will decarbonisation platforms commercialise? How will green steel be sold? How do OpenAI onboard enterprise customers?
Third, a CRM should be built for data and workflows; to help you set objectives, define and execute strategies, and measure outcomes. When you have a CRM that closes this virtuous cycle, then you can become significantly smarter and more effective. Workflow Economics is the glue tying this cycle together, consumer design enables it, and AI will be powered by it.
I’m genuinely interested in these topics and excited about where they are heading, so if you have any thoughts, then please let me know and let’s talk. It’s probably in these discussions that we find the seeds for the next iteration of this thesis.
Effectuation
Being part of building a company reminds me a lot about a concept I (randomly) ended up writing about in my bachelor thesis: effectuation. It’s perhaps unsurprising that I keep getting reminded, since the theory specifically talks about decision-making in entrepreneurial contexts. Either way, it resonated, stuck, and comes back every now and then.
Effectuation states that when making decisions in high-uncertainty contexts, decision-making turns from prediction to action and control. If context is changing fast enough, then don’t spend time trying to analyse what will happen in 5 years, because you can’t predict that - spend time on taking the best available action in front of you. Pick up the shovel and start shovelling. But importantly, with a loose long-term vision in mind. You’re not floating free without purpose, you’re just focused on taking the next best action given a defined direction.
Continuously refining the vision
There are a few very interesting points in the process of building something when the higher-level vision is re-articulated, updated.
It seems likely that these moments are hard to predict, sometimes coming with a flashing insight and other times as a gradual re-definition. These moments often don’t mean a complete reversal vs. the previous vision; rather, they represent a refinement of the existing thesis. In the accounting world, this is called “hardening” - objectives are concretised as a result of learning.
To me, such a moment is deeply personal. Each person has their own long-term vision of where the world is heading, even in an organisation that is highly aligned. Narratives and dreams are always personal, weaving in each person’s unique motivations and understanding of the world. All of this is true in every aspect of life: from personal dreams and decisions, to organisations, to society. The constant is that in all domains, many of us have long-term visions on where we should go, and that vision is continuously refined.
In the case of Planhat, I have previously written about A Data Platform for Action (with Kaveh Rostampor) and about what a Customer Platform is. Each of these were true in the moment and to a large extent still are. But after another year of learning, spending days and nights in the fight, now again I feel is the time to refine and re-articulate the vision. This is by no means the only vision of the future at Planhat, and even though it’s highly aligned with the direction others believe in, it’s also deeply personal in how I choose to express it. That diversity is healthy and welcome.
But hopefully this article can serve as a clear summary to myself, to current and new team members, and to current and new customers, on where Planaht is going.
Big trends
If a trend is big enough, it stays true for a long time.
This is true for the growth of software and digitalisation. For the growth in volume of data generated every single day. And for the growth of businesses moving to more long-term and relationship-focused models; where you are less evaluated on rapid short-term growth than on something sustainable, that can attract and retain and grow customers.
The latter trends serve as the primary need for businesses: how do we become effective and sustainable over the long-term? How do we sell to the right customers, in the right way? How do we then serve, retain and grow them in scalable way? If you can do this, you have a right to exist.
The former trends serve as the input for solving that need: how can you leverage data - especially time-series data - and software to become even more effective?
CRM for Strategies
It seems reasonable that at the intersection of those trends lies something curious.
Software grounded in data that can help you understand who to sell to, how to sell to them, how to then best onboard, support and service them. Crucially, software which houses not only the analysis of data, but also the execution of strategies; when you have both, that’s where virtuous cycles begin.
Such a platform includes a few core parts. You need to 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes.
This is what a CRM should be, and it is - perhaps surprisingly - the most strategic and exciting tool that you have ever seen in business. Building a CRM is building the platform that all the world’s companies will use to deploy data-driven strategies that make them smarter and more effective. How will decarbonisation platforms commercialise? How will green steel be sold? How will OpenAI onboard enterprise customers?
This is actually what our go-to-market teams deal with every day, speaking with leadership at customers on what their biggest commercial priorities are, and help them operationalise those strategies. Do you plan to expand your team next year? Do you plan to move part of your sales organisation to upsell and cross-sell existing accounts? Are you launching a new product? All of the resulting strategies end up in a CRM, because it’s where you can truly operationalise them.
The fascinating part is that given where the trends are, this wasn’t possible to build some years ago, but is now. Data is maturing, and with falling compute costs, processing vast amounts of time-series data is now commercially viable. The speed at which we can build frontend applications is growing by the minute. The growth of Work OS tools (Notion, Monday, Clickup, et al) has brought consumer experiences to enterprise tooling. And at the present moment is - pretentious as it sounds - a generational opportunity to build something incredibly leveraged and impactful to modern business, combining these pieces in thoughtful and creative ways.
To me, this is what Planhat is building: a place where you 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes. Note that this is a natural cycle where your outcomes continuously become your initial customer data input, and your strategies become more optimised as a result of that learning.
This is part data platform, part workflow platform, with the power of the former and intuitive experience of the latter.
Workflow Economics
I am weirdly but proudly passionate about what the market calls Workflows. Workflows are, well, the various flows of work each business runs. You might as well call them processes, but workflows probably sounds a bit more hip.
I am passionate about it because workflows represent the action you take; they represents the forward momentum of your business, and - when perfectly executed - they represent the best version of what you are trying to do. How do you onboard a governmental agency to your security solution? How do you run the design cycle when doing a proof of concept in the electronics hardware industry? Which segment should we go after first when commercialising our new line of industrial robots? What are we doing?
Your workflows, or actions, are what generate outcomes, which is what you care about. Workflows are then combined into overarching strategies that you deploy to reach objectives.
Workflow Economics is a concept that came to mind, which fits very neatly into this cycle. If every business is about running workflows to meet some objectives, doesn’t it make sense to measure which workflows seem to be meeting those objectives and which are not? If you have two ways of onboarding a customer - either self-serve or through a high-touch onboarding - doesn’t it seem pretty critical for you to understand which is more effective at driving adoption, or year 1 renewal (depending on what your objective is)? Or, maybe, if you are meeting all of your objectives (eg, adoption) but you still churn the customer - doesn’t it seem reasonable to understand why adoption is not driving revenue? Similarly, you want to understand the cost side of things, like how much time are we spending with each customer per week?
Workflow Economics is a lens through which you understand the effectiveness and returns of the various processes you run. The processes you run is ultimately who you are as an organisation. So Workflow Economics is what allows you to dissect which parts of your strategy and operations are being successful, and which are not.
Needless to say, Workflow Economics is a core concept to what we are building at Planhat. Planhat is built for you to set objectives, define and execute strategies, and measure outcomes. Workflow Economics is the engine for optimising this cycle.
AI AI AI
Everything today is about AI.
It’s an area that I, like most others, am very curious about - but by far no expert in.
But looking at the intersection of the AI fundamental rules that I do know, and how I think about a CRM for Strategies, you could state some interesting hypotheses.
One such fundamental is that AI is only as good as the data/context it gets. Try prompting ChatGPT, and now try prompting it with 3x the context on the problem, desired outcome, and parameters to consider (or pre-train it on a more relevant dataset). The answer gets significantly better.
It logically follows that the success of AI applications will largely depend on how well you can provide queries with relevant context. By context I mean data on what you are trying to solve, what your desired outcomes are, what has worked well in the past, and what some of the parameters to consider are.
For example, consider prompting some LLM with “what is the next best action to take after this call” and provide it with a summary of a call transcript. Then prompt it with the call transcript, and all the ticket data for the same customer for the last 30 days. Now add all data on how they are using your product. Now add relevant external data, from outside your organisation. And, crucially, now add data on what next best actions has worked historically in similar contexts.
Again, don’t take my word for this - I would love to hear counter-arguments - but it seems logical that, at least up to a certain threshold, accuracy scales with the volume and quality of contextual data.
The implication of this is what? First, I don’t know where the current AI progress will plateau or lead us. Second, knowing that, I bet that a system which is built to consolidate and synthesise all customer data, which contains data on your objectives, and the outcomes of previously executed activities, will be helpful in leveraging AI for benefits.
Architecture
Something which is definitely not my vision, but something I have simply wholeheartedly adopted from our technical founders, is how to architecturally think about building such a software as described above.
It needs to be generic and horisontal. Build modules or components that represent real world concepts, that can be assembled to solve a high variety of problems. And make these pieces interconnected - that is the critical part. Data feeds into segmentation feeds into workflows feeds into analytics and feeds back into another segmentation which feeds into… Close the virtuous cycle of data and action, as measured by workflow economics, and grow.
This is what you see when peeking into Planhat’s technology. An architecture built on data, workflow, automation, and interface platforms. All connected, with a unifying permission and security layer.
The architecture sets the basis for the long-term ability to innovate and expand the product. I am genuinely happy to be working with a team of engineers who have from day 1 built something that I can now be part of innovating on top of.
The fascinating part about effectuation theory is that while it emphasises value in having a clear vision, you’re never any better than what you can cobble together each day. To move an increment forward, knowing vaguely where you want to end up, but with a laser focus on the action. There’s an entirely separate article to be written about culture at Planhat, but this mindset is definitely deeply embedded - don’t get lost in theory, pick up the shovel, and let’s move.
But some key themes have clearly emerged in my thinking, over the past years of building a CRM, that provide overall guidance.
First, effectuation is a real thing. Continuous refinement of vision is good.
Second, building a CRM is the most exciting thing there is, if you think it’s exciting to think about the evolution and operation of modern businesses. How will decarbonisation platforms commercialise? How will green steel be sold? How do OpenAI onboard enterprise customers?
Third, a CRM should be built for data and workflows; to help you set objectives, define and execute strategies, and measure outcomes. When you have a CRM that closes this virtuous cycle, then you can become significantly smarter and more effective. Workflow Economics is the glue tying this cycle together, consumer design enables it, and AI will be powered by it.
I’m genuinely interested in these topics and excited about where they are heading, so if you have any thoughts, then please let me know and let’s talk. It’s probably in these discussions that we find the seeds for the next iteration of this thesis.
Effectuation
Being part of building a company reminds me a lot about a concept I (randomly) ended up writing about in my bachelor thesis: effectuation. It’s perhaps unsurprising that I keep getting reminded, since the theory specifically talks about decision-making in entrepreneurial contexts. Either way, it resonated, stuck, and comes back every now and then.
Effectuation states that when making decisions in high-uncertainty contexts, decision-making turns from prediction to action and control. If context is changing fast enough, then don’t spend time trying to analyse what will happen in 5 years, because you can’t predict that - spend time on taking the best available action in front of you. Pick up the shovel and start shovelling. But importantly, with a loose long-term vision in mind. You’re not floating free without purpose, you’re just focused on taking the next best action given a defined direction.
Continuously refining the vision
There are a few very interesting points in the process of building something when the higher-level vision is re-articulated, updated.
It seems likely that these moments are hard to predict, sometimes coming with a flashing insight and other times as a gradual re-definition. These moments often don’t mean a complete reversal vs. the previous vision; rather, they represent a refinement of the existing thesis. In the accounting world, this is called “hardening” - objectives are concretised as a result of learning.
To me, such a moment is deeply personal. Each person has their own long-term vision of where the world is heading, even in an organisation that is highly aligned. Narratives and dreams are always personal, weaving in each person’s unique motivations and understanding of the world. All of this is true in every aspect of life: from personal dreams and decisions, to organisations, to society. The constant is that in all domains, many of us have long-term visions on where we should go, and that vision is continuously refined.
In the case of Planhat, I have previously written about A Data Platform for Action (with Kaveh Rostampor) and about what a Customer Platform is. Each of these were true in the moment and to a large extent still are. But after another year of learning, spending days and nights in the fight, now again I feel is the time to refine and re-articulate the vision. This is by no means the only vision of the future at Planhat, and even though it’s highly aligned with the direction others believe in, it’s also deeply personal in how I choose to express it. That diversity is healthy and welcome.
But hopefully this article can serve as a clear summary to myself, to current and new team members, and to current and new customers, on where Planaht is going.
Big trends
If a trend is big enough, it stays true for a long time.
This is true for the growth of software and digitalisation. For the growth in volume of data generated every single day. And for the growth of businesses moving to more long-term and relationship-focused models; where you are less evaluated on rapid short-term growth than on something sustainable, that can attract and retain and grow customers.
The latter trends serve as the primary need for businesses: how do we become effective and sustainable over the long-term? How do we sell to the right customers, in the right way? How do we then serve, retain and grow them in scalable way? If you can do this, you have a right to exist.
The former trends serve as the input for solving that need: how can you leverage data - especially time-series data - and software to become even more effective?
CRM for Strategies
It seems reasonable that at the intersection of those trends lies something curious.
Software grounded in data that can help you understand who to sell to, how to sell to them, how to then best onboard, support and service them. Crucially, software which houses not only the analysis of data, but also the execution of strategies; when you have both, that’s where virtuous cycles begin.
Such a platform includes a few core parts. You need to 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes.
This is what a CRM should be, and it is - perhaps surprisingly - the most strategic and exciting tool that you have ever seen in business. Building a CRM is building the platform that all the world’s companies will use to deploy data-driven strategies that make them smarter and more effective. How will decarbonisation platforms commercialise? How will green steel be sold? How will OpenAI onboard enterprise customers?
This is actually what our go-to-market teams deal with every day, speaking with leadership at customers on what their biggest commercial priorities are, and help them operationalise those strategies. Do you plan to expand your team next year? Do you plan to move part of your sales organisation to upsell and cross-sell existing accounts? Are you launching a new product? All of the resulting strategies end up in a CRM, because it’s where you can truly operationalise them.
The fascinating part is that given where the trends are, this wasn’t possible to build some years ago, but is now. Data is maturing, and with falling compute costs, processing vast amounts of time-series data is now commercially viable. The speed at which we can build frontend applications is growing by the minute. The growth of Work OS tools (Notion, Monday, Clickup, et al) has brought consumer experiences to enterprise tooling. And at the present moment is - pretentious as it sounds - a generational opportunity to build something incredibly leveraged and impactful to modern business, combining these pieces in thoughtful and creative ways.
To me, this is what Planhat is building: a place where you 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes. Note that this is a natural cycle where your outcomes continuously become your initial customer data input, and your strategies become more optimised as a result of that learning.
This is part data platform, part workflow platform, with the power of the former and intuitive experience of the latter.
Workflow Economics
I am weirdly but proudly passionate about what the market calls Workflows. Workflows are, well, the various flows of work each business runs. You might as well call them processes, but workflows probably sounds a bit more hip.
I am passionate about it because workflows represent the action you take; they represents the forward momentum of your business, and - when perfectly executed - they represent the best version of what you are trying to do. How do you onboard a governmental agency to your security solution? How do you run the design cycle when doing a proof of concept in the electronics hardware industry? Which segment should we go after first when commercialising our new line of industrial robots? What are we doing?
Your workflows, or actions, are what generate outcomes, which is what you care about. Workflows are then combined into overarching strategies that you deploy to reach objectives.
Workflow Economics is a concept that came to mind, which fits very neatly into this cycle. If every business is about running workflows to meet some objectives, doesn’t it make sense to measure which workflows seem to be meeting those objectives and which are not? If you have two ways of onboarding a customer - either self-serve or through a high-touch onboarding - doesn’t it seem pretty critical for you to understand which is more effective at driving adoption, or year 1 renewal (depending on what your objective is)? Or, maybe, if you are meeting all of your objectives (eg, adoption) but you still churn the customer - doesn’t it seem reasonable to understand why adoption is not driving revenue? Similarly, you want to understand the cost side of things, like how much time are we spending with each customer per week?
Workflow Economics is a lens through which you understand the effectiveness and returns of the various processes you run. The processes you run is ultimately who you are as an organisation. So Workflow Economics is what allows you to dissect which parts of your strategy and operations are being successful, and which are not.
Needless to say, Workflow Economics is a core concept to what we are building at Planhat. Planhat is built for you to set objectives, define and execute strategies, and measure outcomes. Workflow Economics is the engine for optimising this cycle.
AI AI AI
Everything today is about AI.
It’s an area that I, like most others, am very curious about - but by far no expert in.
But looking at the intersection of the AI fundamental rules that I do know, and how I think about a CRM for Strategies, you could state some interesting hypotheses.
One such fundamental is that AI is only as good as the data/context it gets. Try prompting ChatGPT, and now try prompting it with 3x the context on the problem, desired outcome, and parameters to consider (or pre-train it on a more relevant dataset). The answer gets significantly better.
It logically follows that the success of AI applications will largely depend on how well you can provide queries with relevant context. By context I mean data on what you are trying to solve, what your desired outcomes are, what has worked well in the past, and what some of the parameters to consider are.
For example, consider prompting some LLM with “what is the next best action to take after this call” and provide it with a summary of a call transcript. Then prompt it with the call transcript, and all the ticket data for the same customer for the last 30 days. Now add all data on how they are using your product. Now add relevant external data, from outside your organisation. And, crucially, now add data on what next best actions has worked historically in similar contexts.
Again, don’t take my word for this - I would love to hear counter-arguments - but it seems logical that, at least up to a certain threshold, accuracy scales with the volume and quality of contextual data.
The implication of this is what? First, I don’t know where the current AI progress will plateau or lead us. Second, knowing that, I bet that a system which is built to consolidate and synthesise all customer data, which contains data on your objectives, and the outcomes of previously executed activities, will be helpful in leveraging AI for benefits.
Architecture
Something which is definitely not my vision, but something I have simply wholeheartedly adopted from our technical founders, is how to architecturally think about building such a software as described above.
It needs to be generic and horisontal. Build modules or components that represent real world concepts, that can be assembled to solve a high variety of problems. And make these pieces interconnected - that is the critical part. Data feeds into segmentation feeds into workflows feeds into analytics and feeds back into another segmentation which feeds into… Close the virtuous cycle of data and action, as measured by workflow economics, and grow.
This is what you see when peeking into Planhat’s technology. An architecture built on data, workflow, automation, and interface platforms. All connected, with a unifying permission and security layer.
The architecture sets the basis for the long-term ability to innovate and expand the product. I am genuinely happy to be working with a team of engineers who have from day 1 built something that I can now be part of innovating on top of.
The fascinating part about effectuation theory is that while it emphasises value in having a clear vision, you’re never any better than what you can cobble together each day. To move an increment forward, knowing vaguely where you want to end up, but with a laser focus on the action. There’s an entirely separate article to be written about culture at Planhat, but this mindset is definitely deeply embedded - don’t get lost in theory, pick up the shovel, and let’s move.
But some key themes have clearly emerged in my thinking, over the past years of building a CRM, that provide overall guidance.
First, effectuation is a real thing. Continuous refinement of vision is good.
Second, building a CRM is the most exciting thing there is, if you think it’s exciting to think about the evolution and operation of modern businesses. How will decarbonisation platforms commercialise? How will green steel be sold? How do OpenAI onboard enterprise customers?
Third, a CRM should be built for data and workflows; to help you set objectives, define and execute strategies, and measure outcomes. When you have a CRM that closes this virtuous cycle, then you can become significantly smarter and more effective. Workflow Economics is the glue tying this cycle together, consumer design enables it, and AI will be powered by it.
I’m genuinely interested in these topics and excited about where they are heading, so if you have any thoughts, then please let me know and let’s talk. It’s probably in these discussions that we find the seeds for the next iteration of this thesis.
Effectuation
Being part of building a company reminds me a lot about a concept I (randomly) ended up writing about in my bachelor thesis: effectuation. It’s perhaps unsurprising that I keep getting reminded, since the theory specifically talks about decision-making in entrepreneurial contexts. Either way, it resonated, stuck, and comes back every now and then.
Effectuation states that when making decisions in high-uncertainty contexts, decision-making turns from prediction to action and control. If context is changing fast enough, then don’t spend time trying to analyse what will happen in 5 years, because you can’t predict that - spend time on taking the best available action in front of you. Pick up the shovel and start shovelling. But importantly, with a loose long-term vision in mind. You’re not floating free without purpose, you’re just focused on taking the next best action given a defined direction.
Continuously refining the vision
There are a few very interesting points in the process of building something when the higher-level vision is re-articulated, updated.
It seems likely that these moments are hard to predict, sometimes coming with a flashing insight and other times as a gradual re-definition. These moments often don’t mean a complete reversal vs. the previous vision; rather, they represent a refinement of the existing thesis. In the accounting world, this is called “hardening” - objectives are concretised as a result of learning.
To me, such a moment is deeply personal. Each person has their own long-term vision of where the world is heading, even in an organisation that is highly aligned. Narratives and dreams are always personal, weaving in each person’s unique motivations and understanding of the world. All of this is true in every aspect of life: from personal dreams and decisions, to organisations, to society. The constant is that in all domains, many of us have long-term visions on where we should go, and that vision is continuously refined.
In the case of Planhat, I have previously written about A Data Platform for Action (with Kaveh Rostampor) and about what a Customer Platform is. Each of these were true in the moment and to a large extent still are. But after another year of learning, spending days and nights in the fight, now again I feel is the time to refine and re-articulate the vision. This is by no means the only vision of the future at Planhat, and even though it’s highly aligned with the direction others believe in, it’s also deeply personal in how I choose to express it. That diversity is healthy and welcome.
But hopefully this article can serve as a clear summary to myself, to current and new team members, and to current and new customers, on where Planaht is going.
Big trends
If a trend is big enough, it stays true for a long time.
This is true for the growth of software and digitalisation. For the growth in volume of data generated every single day. And for the growth of businesses moving to more long-term and relationship-focused models; where you are less evaluated on rapid short-term growth than on something sustainable, that can attract and retain and grow customers.
The latter trends serve as the primary need for businesses: how do we become effective and sustainable over the long-term? How do we sell to the right customers, in the right way? How do we then serve, retain and grow them in scalable way? If you can do this, you have a right to exist.
The former trends serve as the input for solving that need: how can you leverage data - especially time-series data - and software to become even more effective?
CRM for Strategies
It seems reasonable that at the intersection of those trends lies something curious.
Software grounded in data that can help you understand who to sell to, how to sell to them, how to then best onboard, support and service them. Crucially, software which houses not only the analysis of data, but also the execution of strategies; when you have both, that’s where virtuous cycles begin.
Such a platform includes a few core parts. You need to 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes.
This is what a CRM should be, and it is - perhaps surprisingly - the most strategic and exciting tool that you have ever seen in business. Building a CRM is building the platform that all the world’s companies will use to deploy data-driven strategies that make them smarter and more effective. How will decarbonisation platforms commercialise? How will green steel be sold? How will OpenAI onboard enterprise customers?
This is actually what our go-to-market teams deal with every day, speaking with leadership at customers on what their biggest commercial priorities are, and help them operationalise those strategies. Do you plan to expand your team next year? Do you plan to move part of your sales organisation to upsell and cross-sell existing accounts? Are you launching a new product? All of the resulting strategies end up in a CRM, because it’s where you can truly operationalise them.
The fascinating part is that given where the trends are, this wasn’t possible to build some years ago, but is now. Data is maturing, and with falling compute costs, processing vast amounts of time-series data is now commercially viable. The speed at which we can build frontend applications is growing by the minute. The growth of Work OS tools (Notion, Monday, Clickup, et al) has brought consumer experiences to enterprise tooling. And at the present moment is - pretentious as it sounds - a generational opportunity to build something incredibly leveraged and impactful to modern business, combining these pieces in thoughtful and creative ways.
To me, this is what Planhat is building: a place where you 1) consolidate and synthesise all commercial data, 2) set objectives, 3) define and execute strategies, and 4) measure outcomes. Note that this is a natural cycle where your outcomes continuously become your initial customer data input, and your strategies become more optimised as a result of that learning.
This is part data platform, part workflow platform, with the power of the former and intuitive experience of the latter.
Workflow Economics
I am weirdly but proudly passionate about what the market calls Workflows. Workflows are, well, the various flows of work each business runs. You might as well call them processes, but workflows probably sounds a bit more hip.
I am passionate about it because workflows represent the action you take; they represents the forward momentum of your business, and - when perfectly executed - they represent the best version of what you are trying to do. How do you onboard a governmental agency to your security solution? How do you run the design cycle when doing a proof of concept in the electronics hardware industry? Which segment should we go after first when commercialising our new line of industrial robots? What are we doing?
Your workflows, or actions, are what generate outcomes, which is what you care about. Workflows are then combined into overarching strategies that you deploy to reach objectives.
Workflow Economics is a concept that came to mind, which fits very neatly into this cycle. If every business is about running workflows to meet some objectives, doesn’t it make sense to measure which workflows seem to be meeting those objectives and which are not? If you have two ways of onboarding a customer - either self-serve or through a high-touch onboarding - doesn’t it seem pretty critical for you to understand which is more effective at driving adoption, or year 1 renewal (depending on what your objective is)? Or, maybe, if you are meeting all of your objectives (eg, adoption) but you still churn the customer - doesn’t it seem reasonable to understand why adoption is not driving revenue? Similarly, you want to understand the cost side of things, like how much time are we spending with each customer per week?
Workflow Economics is a lens through which you understand the effectiveness and returns of the various processes you run. The processes you run is ultimately who you are as an organisation. So Workflow Economics is what allows you to dissect which parts of your strategy and operations are being successful, and which are not.
Needless to say, Workflow Economics is a core concept to what we are building at Planhat. Planhat is built for you to set objectives, define and execute strategies, and measure outcomes. Workflow Economics is the engine for optimising this cycle.
AI AI AI
Everything today is about AI.
It’s an area that I, like most others, am very curious about - but by far no expert in.
But looking at the intersection of the AI fundamental rules that I do know, and how I think about a CRM for Strategies, you could state some interesting hypotheses.
One such fundamental is that AI is only as good as the data/context it gets. Try prompting ChatGPT, and now try prompting it with 3x the context on the problem, desired outcome, and parameters to consider (or pre-train it on a more relevant dataset). The answer gets significantly better.
It logically follows that the success of AI applications will largely depend on how well you can provide queries with relevant context. By context I mean data on what you are trying to solve, what your desired outcomes are, what has worked well in the past, and what some of the parameters to consider are.
For example, consider prompting some LLM with “what is the next best action to take after this call” and provide it with a summary of a call transcript. Then prompt it with the call transcript, and all the ticket data for the same customer for the last 30 days. Now add all data on how they are using your product. Now add relevant external data, from outside your organisation. And, crucially, now add data on what next best actions has worked historically in similar contexts.
Again, don’t take my word for this - I would love to hear counter-arguments - but it seems logical that, at least up to a certain threshold, accuracy scales with the volume and quality of contextual data.
The implication of this is what? First, I don’t know where the current AI progress will plateau or lead us. Second, knowing that, I bet that a system which is built to consolidate and synthesise all customer data, which contains data on your objectives, and the outcomes of previously executed activities, will be helpful in leveraging AI for benefits.
Architecture
Something which is definitely not my vision, but something I have simply wholeheartedly adopted from our technical founders, is how to architecturally think about building such a software as described above.
It needs to be generic and horisontal. Build modules or components that represent real world concepts, that can be assembled to solve a high variety of problems. And make these pieces interconnected - that is the critical part. Data feeds into segmentation feeds into workflows feeds into analytics and feeds back into another segmentation which feeds into… Close the virtuous cycle of data and action, as measured by workflow economics, and grow.
This is what you see when peeking into Planhat’s technology. An architecture built on data, workflow, automation, and interface platforms. All connected, with a unifying permission and security layer.
The architecture sets the basis for the long-term ability to innovate and expand the product. I am genuinely happy to be working with a team of engineers who have from day 1 built something that I can now be part of innovating on top of.
The fascinating part about effectuation theory is that while it emphasises value in having a clear vision, you’re never any better than what you can cobble together each day. To move an increment forward, knowing vaguely where you want to end up, but with a laser focus on the action. There’s an entirely separate article to be written about culture at Planhat, but this mindset is definitely deeply embedded - don’t get lost in theory, pick up the shovel, and let’s move.
But some key themes have clearly emerged in my thinking, over the past years of building a CRM, that provide overall guidance.
First, effectuation is a real thing. Continuous refinement of vision is good.
Second, building a CRM is the most exciting thing there is, if you think it’s exciting to think about the evolution and operation of modern businesses. How will decarbonisation platforms commercialise? How will green steel be sold? How do OpenAI onboard enterprise customers?
Third, a CRM should be built for data and workflows; to help you set objectives, define and execute strategies, and measure outcomes. When you have a CRM that closes this virtuous cycle, then you can become significantly smarter and more effective. Workflow Economics is the glue tying this cycle together, consumer design enables it, and AI will be powered by it.
I’m genuinely interested in these topics and excited about where they are heading, so if you have any thoughts, then please let me know and let’s talk. It’s probably in these discussions that we find the seeds for the next iteration of this thesis.
Daniel Sternegard
•
Director of Product, Planhat
Daniel leads Planhat's overall product strategy and roadmap. Prior to Planhat, Daniel was at McKinsey & Company, primarily focused on corporate and commercial strategy in Advanced Industries.
Don't miss these
Guide
•
6 minute read
Multi-Year Deals: The basics
Christian Jakenfelds
Guide
•
9 minute read
9 customer success strategies to reduce churn
Jonas Terning
Guide
•
7 minute read
Top 9 most important customer success KPIs
Jonas Terning
Don't miss these
Guide
•
6 minute read
Multi-Year Deals: The basics
Christian Jakenfelds
Guide
•
9 minute read
9 customer success strategies to reduce churn
Jonas Terning
Guide
•
7 minute read
Top 9 most important customer success KPIs
Jonas Terning
Don't miss these
Guide
•
6 minute read
Multi-Year Deals: The basics
Christian Jakenfelds
Guide
•
9 minute read
9 customer success strategies to reduce churn
Jonas Terning
Guide
•
7 minute read
Top 9 most important customer success KPIs
Jonas Terning
Don't miss these
Guide
•
6 minute read
Multi-Year Deals: The basics
Christian Jakenfelds
Guide
•
9 minute read
9 customer success strategies to reduce churn
Jonas Terning
Guide
•
7 minute read
Top 9 most important customer success KPIs
Jonas Terning
Thought-leading customer-centric content, direct to your inbox every month.
By submitting this form I agree that Planhat may collect, process and retain my data pursuant to its Privacy Policy.
Customers
© 2024 Planhat AB
Thought-leading customer-centric content, direct to your inbox every month.
By submitting this form I agree that Planhat may collect, process and retain my data pursuant to its Privacy Policy.
Customers
© 2024 Planhat AB
Thought-leading customer-centric content, direct to your inbox every month.
By submitting this form I agree that Planhat may collect, process and retain my data pursuant to its Privacy Policy.
Customers
© 2024 Planhat AB
Thought-leading customer-centric content, direct to your inbox every month.
By submitting this form I agree that Planhat may collect, process and retain my data pursuant to its Privacy Policy.
Customers
© 2024 Planhat AB