
The foundation of customer and product discovery hinges on two fundamental questions: What do we aim to learn, and who can provide us with that knowledge?

This is why discovery in the B2B context is often considered more complex. Instead of selling to just one person, you're dealing with a collective—a company—ensuring your product adds value to the group as a whole. This can lead to decision fatigue for teams embarking on the discovery process as they don't know who they start with interviewing.

In general, product development requires consideration of three key groups: decision-makers, champions, and other stakeholders.

The decision-maker is the individual who has the authority to decide whether a solution should be purchased and implemented within a company. This is typically a higher level manager or in smaller companies the C-Suite. It's obvious why the decision maker is important for your products success: Without her support, you can't succeed.

The champion is the person who deeply understands the problems you want to solve with your product. They have the time and motivation to figure out a solution - like your product - and advocate to buy it. Sometimes the champion got the explicit order from the decision maker to search for a solution, sometimes the champion is the driving force, who has to pitch in front of the decision maker. The champion is important to your product, as she understands the problem space the most and has probably the most motivation to talk to you about it.

Last and least there are the other stakeholders. They influence the problem space of the champion and decision maker and can halt the decision making process to buy a new product. You can't afford to ignore them, but when you start out, they are definitely not the focal point of your discovery.

In small companies, the decision-maker and champion might be the same person. But when that's not the case, a crucial question arises:

Should our discovery focus be on the decision-maker or the champion?

Generally, your discovery work should be steered by your most significant risks—with a goal to mitigate them. When starting out, if uncertainty looms, it's wise to focus on the champion. The role of the champion is to drive his company to buy and use the solution, the role of the decision maker is more of a gate-keeper, which has to be overcome. This champion not only understands the problem better but is also more approachable. The insights gained from her can subsequently pave the way for easier access to the actual decision-maker.

## Champions have a deep understanding of the problem

The champion is by definition the person who is in search of a solution.

The champion's role involves finding a solution—a task explicitly assigned by a manager, a responsibility inherently tied to her role, or a challenge she is intrinsically motivated to tackle. But let's use an example. One of the offers of my team is to do product discovery coaching (meta, right?). So the three options might play out as:

Explicit task from manager: "Make sure your team does good product discovery" or even "Search for a product discovery coach for your team"
Results from her responsibilities: The champion is responsible for the product discovery practice of this team.
Intrinsic motivation: The champion is the manager of the team and wants to introduce product discovery, but lacks time or skills to coach them.

In each case, the champion thinks through the problem space and the different options they have to tackle resolve that problem.

This engagement results in a deep understanding of the specific challenges her company faces when applying product discovery, rather than a broad understanding of product discovery in general. She assesses the core challenges the company faces, connects with various stakeholders, and gathers their perspectives on overcoming these hurdles.

This means not only, that the champion is your ally in solving the problem, she already collected valuable insights you want to collect in your discovery: What triggered the company's awareness of the problem, and how is the company framing this issue? What are the different needs and desires in the process of overcoming the problem? What are the constraints different stakeholders place on the solution?

The champion will help you gather the biggest part of the data you need.

## Champions are more approachable

We've already discussed that champions are responsible to understand the specific challenges for their company and find a solution to it.

Sometimes champions invest a substantial amount of time into understanding these challenges and finding solutions, while other times they may neglect this responsibility due to being overwhelmed with other tasks. Either way, they are interested in any assistance that can help them understand the challenge, explore different options to solve it, and determine the criteria they should use to evaluate a solution.

This is exactly what you can offer champions, when you do discovery with them.

In an interview, you use your domain expertise to ask questions that shed light on previously overlooked aspects and criteria of their challenge. While you concentrate on past behavior in your interviews to overcome psychological biases, you often connect the dots for them. You can also reserve some time at the end of the interviews to share what you already have learned about this problem space and give a glimpse on your solution, which offers a new way of working where some of their needs, desires or problems are taken care of.

Champions care deeply about the problem you solve, they'll help you so you can help them.

## The insights from the champions lead to easier access to the decision makers

To round this up, the question to focus on champions vs decision-makers is so hard, because we have to satisfy both needs to be successful with our product.

However, the champion plays a more central role in our discovery process and is typically more approachable. She also gives us the insights we need to do discovery with the decision-maker. Unless you have connections in your target industry, it's harder to approach decision-makers. The problem we solve is typically a smaller part of their responsibilities compared to those of the champion. This means our messaging, when reaching out to them, must be much more specific to appear relevant. This will show them that we've done our homework and that we respect their time.

My point of view here is supported by marketing and positioning experts as well.

In early product discovery what we learn is the foundation of a good positioning. April Dunford spoke in her new podcast "Positioning with April Dunford" in the episode [Single-Product vs. Multi-Product Positioning: A Comprehensive Guide](https://www.positioning.show/single-product-vs-multi-product-positioning/), about the importance of the champion for the effective positioning.

> For positioning work, the champion matters like literally times a thousand versus everybody else. [...] The champion is the only one who can make the deal happen. Everybody else has the ability to kill the deal.

And if your organization relies on Inbound-Marketing, I have two other thought-leaders for you. Brian Halligan and Dharmesh Shah emphasize in their book "Inbound Marketing":

> To be successful and grow your organization, you need to match the way you market your products to the way prospects learn about and shop for your products. And you do that by generating leads through inbound marketing.

For effective inbound marketing you need to know how your prospects learn about your product. And who does this? The champion.


The foundation of customer and product discovery hinges on two fundamental questions: What do we aim to learn, and who can provide us with that knowledge?
This is why discovery in the B2B context is often considered more complex. Instead of selling to just one person, you’re dealing with a collective — a company — ensuring your product adds value to the group as a whole. This can lead to decision fatigue for teams embarking on the discovery process as they don’t know who they start with interviewing.

Champion Vs. Decision Maker: Whom to Target in B2B Customer Discovery?


The discussion surrounding outcomes over outputs is crucial in the shift towards a product-led organization. In traditional project management, teams focus on deliverables like MVPs, features, sales slide decks, or webinars and optimize for their swift delivery.

However, transitioning as an organization to a new mindset requires some individuals to become evangelists for the new approach - in this case, thinking about outcomes, aka what we really try to achieve.

Learning this new way of thinking is incredibly challenging, and numerous forces of inertia within organizations can hinder these evangelists. Understandably, they become frustrated and fixate on the promised land the product thought leaders have taught them. This can lead to difficult and even emotional discussions between them and the rest of the organization.

In my experience, these discussions become very semantical. People may start to argue what the terms output, outcomes, and impact mean and whether a specific goal is an outcome or not. Then a discussion kicks off if we must define an outcome in the first place.

Instead, we should engage in more goal-oriented discussions.

In this article, we will dig into the first principles of goal-setting, which are the mechanics behind this semantic discussion of outcomes vs. output. We will foster alignment in the team on the outcome and the output, establish a metric that guides our work, and commit to clear actions.


## Aligning on Success Criteria Without Criticizing Ideas

The first principle is the core reason why a shift to outcome thinking is so important and will sound familiar to you because this might be the discussion you are already having. Let us start with an example:

We are a team of product coaches and want to organize a webinar about "Outcomes over Output".

> **The action we plan**: Organizing a webinar about "Outcomes over Output" for product managers in 4 weeks.

One of our team members - Steve - has been reading Marty Cagan, though. He notes that we just set an output goal. We need to define outcome goals; the best webinar is useless if no one shows up.

The team gets defensive and insecure. How can they commit to an outcome like "generated leads through the webinar" or the number of attendees if it was so hard to do it by organizing a webinar?

While Steve is right in theory, the team is stuck and is wasting a lot of energy in theoretical debates and emotional discomfort. Let us help him shift the discussion to a more result-oriented perspective.

Let us step back and have a look at the first principles of this problem: The team is trying to solve a complex and ill-defined problem. They expect two effects from their action:
- We expect that people will attend.
- We expect that our business will benefit from it.

The thing is: it is almost impossible to predict these two things when you start out and have not done any webinars for this target group.

Stephen Bungay identified in his book "The Art of Action" these three gaps when we plan for an ill-defined complex problem:

- Knowledge Gap: We cannot know everything about our customers
- Alignment Gap: There is always a gap between what we want our team members to do and what they do
- Effect Gap: We cannot predict how customers will react

There are two levers to fight these gaps: First, we need to define the outcome/effect we want to achieve. Second, we must give the teams permission to learn through early feedback from real customers and adapt. They must have permission to deviate from the original implementation idea.

But when we challenge the ideas of people, they can get offended easily.

So acknowledging that we cannot reliably plan how to achieve the desired effects (outcomes), we challenge the ideas of our coworkers. We ask why they think something would be a good idea. We want to know how they can be so sure that this webinar with this topic will lead to the desired results.

They get protective of their ideas instead of listening to you.

We don't want their guards up, so we need to make them feel safe. Make sure that this discussion is not to challenge their idea, but to be aligned on all levels. What are our ambitions with this webinar, and what people do we expect to attend?

When are you proud of our achievements when we organize this webinar? How does success look like for us?


> **The action we plan**: Organizing a webinar about "Outcomes over Output" for product managers in 4 weeks.
>
> **Success Criteria**: 40 attending Product Managers - by attending to the webinar we get a signal, that this topic is relevant to them.

Now they feel heard and share their valuable thoughts. You don't come across as pushy but interested and curious.

So, with this step, we have established an idea of what we want to do and what we want to achieve.

## Establishing Learning Metrics to Guide Our Work

We are aligned on what we want to do and what results we expect from it.

We already know that the link between what we do and what we want to achieve (Success criteria) is hard to plan in advance. It's often determined by how we do what we do.

So how can we learn how our webinar should be marketed, structured, and how the exact target group looks like?

Our measurement of success (40 product managers attend) is something we measure after the webinar has been conducted; this is too late. We market the webinar, prepare it, and host it, and afterward, we know how many have attended.

This means our current goal cannot guide our learnings.

So let's define a measurement we can observe while we are preparing the webinar. This is something experienced marketers and product managers will do automatically, but this step ensures the team aligns beforehand on their ambitions and understanding of the problem area.

To do that, let's just ask: What can we observe while we work, which is a predictor of our success of 40 product managers attending?

If your team is overwhelmed by this question, think about the customer journey. What customer behavior can we observe, which must happen (and therefore is an indicator) so 40 product managers will attend?

1. X Product Managers become aware
2. X product managers register for the webinar
3. 40 product managers attend the webinar

The critical step here for us is registration. This is something we can observe while we work, and it is tightly linked to our success criteria of 40 attendees. So how many product managers have to register so 40 of them will attend? We can calculate backward. With a show-up rate of 40%, we need 100 webinar registrations.

Our webinar is supposed to take place in 4 weeks. So we need 25 registrations per week.

> **The action we plan**: Organizing a webinar about "Outcomes over Output" for product managers in 4 weeks.
>
> **Success Criteria**: 40 attending Product Managers - by attending to the webinar we get a signal, that this topic is relevant to them.
>
> **Learning Metric**: 100  registered product managers (25 per week)

This is a metric that can guide our work, and we can learn if what we do is sufficient to reach the target. Just to give you a backlink to the theoretical discussions: The learning metric is the leading indicator to the lagging success criteria - both are outcomes.

## Committing to Clear Actions and Learning Which Drive Goal Achievement

The team is aligned on what success looks like and how we can learn if what we're doing has the desired effect.

When we have a lofty outcome, teams can still struggle to commit to it.

What do I mean by lofty outcomes? Our team has never organized or marketed a webinar. Now they know how they can learn if they're on the right track, but they still don't know how to acquire participants. The outcome doesn't give them enough structure to know what to do. We have barely discussed what it takes to attract attendees. So, how can they even commit to it? What if they don't make it?

First, we have to acknowledge the emotions this requested commitment elicits.

Most people aren't used to committing to outcomes. Committing to an outcome means something different than committing to delivering a webinar. If you fail to deliver a webinar, you've fallen short. This is "solely" your responsibility.

If you don't achieve an outcome, that's okay. Nobody can guarantee it. Committing means we achieve it, and if we don't, we learn what questions we should have been asking in the planning phase (e.g., how can we narrow down the target group?).

But this is scary, and people are uncertain about what they are committing to, so what helps in this situation is aligning on clear actions initially.

To do that, we brainstorm what actions you can take to achieve your learning goal of 25 registrations per week over 4 weeks (and eventually 40 product managers attending). You are looking for different strategies to evolve. Now you can choose the 2 or 3 best and try them and learn how they perform.

First, brainstorm strategies without any metrics to speed up the process (you don't have to argue about the exact number in a strategy you won't choose):

- We can message X product managers on LinkedIn, and a Y% of them respond.
- We can place an ad that converts X product managers to visit the landing page.
- We can invite X of our customers via email.
- We can share the webinar in our newsletter with X product managers.

Second, we can choose 2-3 of these strategies and commit to them for the first week. Then we can measure the impact of the strategies and learn: Is one of these strategies sufficient to achieve our ultimate goal of 100 registrations? Which one is the most effective?

The team can commit to clear actions and learning about a hypothesis. These are actions they can own!


Discussions about outcomes and outputs often become very semantical.

People may start to argue what the terms output, outcomes, and impact mean and whether a specific goal is an outcome or not. Then a discussion kicks off if we must define an outcome in the first place.

Nobody is talking about the actual goals.

So we need to shift the discussions to a more goal-oriented perspective.

Effective Goal-Setting without the Semantical Debate of Outcomes vs. Outputs



The process of defining your customer segments can feel cumbersome, when you want to develop your product.

It certainly did for me. Plus, there is so much wrong about how customer segments and personas are used. You can easily get lost in different kinds of demographics like age and gender or firmographics like number of employees or annual recurring revenue.

You might wonder how customer segments and personas relate. Personas are exemplary representatives of the customer segment to be able to build more empathy. You give them a name and an avatar to reference them easier.

This article is about the importance of a good customer segment definition.

## Why we need customer segments in the first place

To evaluate how good the criteria are that we use to define our customer segments, we first need to know what customer segments are there for.

When we create a new product, we first make sure, that customers become aware of if and choose it over its alternatives.

So, our first challenge is our fight for attention. We can win this fight with a very specific marketing message. It must be easy to understand and highly relevant. We can achieve this by understanding the problem of our customers well and getting specific enough with our targeting.

Speaking of targeting: In order for customers to perceive the message, we must first bring it in front of them. We need to know which criteria we can use to target them.

Last, but definitely not least, our product must create more value than the competition. As there is a solution to almost every problem out there, it becomes harder and harder to build the best one. The last resort is to better solve the more specific problem of a smaller submarket.

To be successful as a company marketing and creating this product, we must be on the same page. This is where a well-defined customer segment comes to the rescue. Defining these customer segments poorly can lead to very weird discussions about what our customers need.

When two people think of different customers and argue about their problems, a lot of emotions can come up.


## Defining Customer Segments with irrelevant criteria leads to bad results

Customer segments are defined by different criteria. One famous example on the internet is this:

Our target group is:

- male
- Born in 1948
- Raised in the UK
- Married Twice
- Lives in a castle
- Wealthy and famous

This is a *very specific* customer segment, isn’t it? Well, here are two representatives of this segment:

![King Charles & Ozzy Osbourne](/imgs/prince-charles-and-ozzy-osbourne.jpeg)

This is nuts! Prince Charles and Ozzy Osbourne?

How can we craft a marketing message, which resonates with these two guys? How can we tailor a product of higher value than the existing alternatives to their needs?

This sounds like a bad customer segment definition, and for many products it is. People in this customer segment can have so different world views, lifestyles, and jargons.

To evaluate how good the customer segment definition is, we need to put it in relation to a product
## Customer segments have to be evaluated in context

While Charles and Ozzy seem to have nothing in common, there might be products which are highly relevant to both of them. They are older males who have enjoyed British food for over 70 years. And they are quite solvent.

So when our product is a new luxury haemorrhoids cream, this segment might hit close home.

> Rapid relief of pain & itching – Enjoy your second life in your castle again!

There are other products or services, where this customer segment isn’t as bad as it sounded without any context. E.g., a divorce lawyer or British country style gun cabinets…

And I will add that Ozzy Osbourne will stand out in any customer segment.

Nevertheless, what can we learn from it?

## Evaluating the quality of your customer segment criteria

Remember the purpose of the customer segments? Highly relevant to our customers. Who are our customers? The people to whom we can add the most value (and who are willing to pay for it).

These criteria just help to “target” them. This means each of the criteria has to correlate with the problem we solve (or with the desire, we nurture) or other parts of our business model.

Let’s check our example of the haemorrhoid cream:

- male → probably relevant for marketing messaging
- born in 1948 → while maybe too specific, older people are more prone to developing haemorrhoids.
- Raised in the UK → this sounds quite cultural, but I guess living in the UK is relevant for distribution of your product
- Married Twice → Did they have a stressful life?
- Lives in a castle → probably irrelevant and should be deleted, unless it’s cold there
- Wealthy and Famous → They have money for a luxury brand and might pay extra for fast results

We should alter the criteria for our customer segment here. But one thing is notable here: There doesn’t have to be a causation between the business model and the criteria itself. A correlation is sufficient, as they not only help with empathy, but with targeting too.

This leads to the last requirement for our criteria, some of them must be actionable. In order to target customers, e.g., in a marketing campaign or for outbound sales on LinkedIn, we need criteria to filter for our target market.

## Double checking your critera

To summarise, you can check your criteria with the following questions:

- Does every criterion at least correlate with your business model (problem you solve, obstacles people have, distribution channel you want to use)
- Are there multiple customers in your segment with vastly different problems? Aren’t they outliers like Ozzy?
- Can we define additional criteria to create a segment with the majority being our target customers?

Often times we have the urge that our customer segment is too small. More often than not, this is not the case. Your product and marketing messaging will resonate with people outside your customer segment.


Defining good customer segments is not easy. And teams can get lost in it.
So it is crucial to know what is important in customer segments and what criteria are suitable to define them.

What are good Criteria for Defining Customer Segments?



# My 2022 review and lookout

2022 was a great year for me.

It was exhausting, but also very rewarding. I got to do interviews with a lot of product owners and product managers about the current conditions of their product management practices. Also, I was able to have a little impact on the German product community by hosting 11 product lunches this year.

And that happened while my son started walking. Nowadays, he even babbles incessantly about fire departments, cranes, and excavators. Just unbelievable.

Aside from that, there are a lot of things I can proudly look back to:

- I’ve written 130 posts in the past 134 days
- Josua and I have established a new webinar format Product Lunch
- Conducted way over 50 interviews with product people from Germany
- Closed 2 entirely new clients
- Have over 10 prospects in our pipeline
- I’ve narrowed down what I want to write about: strategic clarity
- My first big public talk at Digitale Leute Summit about business experiments

I've taken on a lot of responsibilities this year, and I wish I had done a better job of most of them.

This is what the yearly review is about, isn’t it?

## Shocker: Not time is my scarcest resource – energy is.

I’ve been struggling with my to-do list for quite some time now.

But this year I improved my consciousness about the root causes a lot. The first epiphany was, that it’s not only a problem of finite time. It’s that I just have 2 or 3 hours a day, where I can really push forward in deep work.

So planning 8 hours of hard thinking just leads to unrealistic expectations and ultimately stress.

I don’t finish the things I do, and I get sloppy. So, my review questions have changed to:

- What did I enjoy?
- What gave me Energy?
- What drained my Energy?

They help me to realise that there is very limited room for fields I can grow in.

In the past, I’ve always learned by overselling myself. I took on responsibilities for things I wasn’t capable of solving yet. So, I pushed, learned, and accomplished it somehow.

But this only works for 1 to max 2 things in parallel – not 5.

## Assessment of the last year

The main thing, which drained my energy this year, was missing focus.

I've taken on too many new tasks where I don't have the skills I want to have to meet my standards. As mentioned earlier, taking on just one of these responsibilities leads to challenging and good growth.

I did it with these 5 to different degrees:

- **Leading my consulting team as its Head of**
- **Doing discovery for our consulting services** \[in a very demanding setting because of non-obvious customer segments with different levels of problem-awareness]
- **Selling consulting services** \[in B2B\]
- **Consult my clients** in finding viable business models, starting with product discovery and developing a strategy
- **Starting to write**

The results by which I am ultimately measured are sales numbers.

On one hand, I was lucky, as sales comes relatively easy to me and the challenge at hand isn’t as complex and hard-to-navigate as the discovery or client work.

But I have to get familiar with a new process. Juggling all the different sales opportunities, documenting them and following up on everybody was very stressful for me. I expect this to become an easy and automatic process in the future, but initially, it’s quite energy consuming.

The processes I have to establish:

- Juggling all the sales opportunities, documenting them and following up on everybody
- Documenting my work, so I can communicate clearly what I did to stakeholders
- Document my strategic status quo to give context to my team
- Write posts in a clear manner

So in the future, I want to acknowledge and limit the number of complex challenges I tackle simultaneously. And establishing a new process counts as a complex challenge.

Having that said, I realised again in my reflection, what gives me energy and drives my work: Creating strategic clarity.

I love being confident, that what I do is contributing to a big and meaningful goal. And I love doing that for my clients as well. Whether it's pointing out the weaknesses in their unit economics or telling them to focus on another part of their business.

And many customers are just as grateful as I am for this outside view of their challenges at the strategic level.

## My Resolutions

In 2023, I want to be clearer about my challenges at the various strategic levels. To this end, I will use Martin Erkisson’s [decision stack](https://www.thedecisionstack.com/).

I also intend to keep writing and find an even easier mode with less friction to invest my energy in other things. Plus, I would like to create more blog posts as they are less ephemeral than posts on Twitter and LinkedIn.

I would like to expand my relationship with the German product community, as I really enjoy interacting with this diverse group of people.

Finally, I want to put less pressure on myself. Much of that pressure comes from the desire to buy a house in 4 years. Instead, I intend to give myself some air and grow next year from the opportunities I face at tarent right now.

This means in Q1 of 2023 I will focus on my sales skills and processes to broaden my network and help more people identify what potential modern product management can have in their businesses.

On the other hand, I stay active in leading the consulting team, in discovery, and in writing, but I don't aspire to grow to the next level. This is my conscious prioritisation of complex challenges to tackle.

I will never compromise my client's work.


2022 was a great year for me.
It was exhausting, but also very rewarding. And it lead to an epiphany regarding time management. What I did do wrong and how I might do better. 

My 2022 review


The mission of our team is to help agile software development teams building the right things from an entrepreneurial perspective. Our background is not in “product management”, but in start-ups.

While there is a big overlap between the two, my experience with full-fledged product discovery is pretty limited. I’ve been helping teams in bigger companies adopt discovery for a year now.

This is not much, but I’ve conducted over 50 interviews, too. This gave me the opportunity to study many teams with an outside perspective. Many of them appreciated this outside perspective, as they’ve been caught up in the daily routine.

Yes, you got it right, we do discovery interviews about discovery – Meta, right?

You might think, why this story is relevant to you? Our first goal was to introduce a continuous discovery approach into companies, so they figure out the right things to build. But we learned this might not be the best first step.

If you want to introduce discovery in your company, you might get a better perspective what you should focus on.

## Do teams need help in prioritising the right features or gather data to make an informed decision?

We started with the research goal of how product teams are prioritising what to build today. Our hypothesis was, that they had no data to make informed decisions.

Our research has confirmed our hypothesis. The problem is, that we formulated a hypothesis prone to confirmation bias. Product owners didn’t have any data to prioritise right, but it wasn’t a concern to them.

Their #1 problem is stakeholder management.

Their de facto job is gathering the needs and wants of different stakeholders. Then they have to build consensus between them and map that consensus to the backlog.

The product owners have little say in the discussion about what to build.

## What is their goal?

Glad we realised quite fast, that we were asking the wrong question.

So we shifted our research focus to the goal of the team. Our next hypothesis was: We can show teams that they can improve what to build to have a bigger impact on their goal. But asking for their goals was pretty hard.

We tried different framings:

- What is the current goal your team tries to accomplish?
- What does success look like for your team?
- How is your performance assessed by management?

We struggled to start a valuable discussion with these questions.

Turns out, most companies neither have a good strategy nor good goals defined. The team's performance was assessed on velocity of features shipped and a subjective assessment of how good the product was. When the business metrics like revenue over all teams weren’t healthy, pressure increased to ship more or better features.

So, we shifted our research to defining strategy and goals to have a basis for product management.

## Helping the teams define their goals

Research goal #3: How are companies defining their strategy and goals?

This question is a tough nut to crack. Our network consists mostly of individual contributor (IC) but not product leaders. Yet, most of this process happens on the leadership level. Talking to the IC level was still helpful.

The product teams had little say in the strategy.

The leadership communicated mainly features to build as the result of the strategy. And they didn’t ask for any input. Speaking to leaders, there was often no problem-awareness.

Problem-awareness is necessary for us to help these teams. So, our research shifted another time.

## Symptoms of missing strategy

Helping teams with strategy is such an unspecific and broad promise.

So we need to understand the problems a missing strategy causes. One obvious to us is missing focus in your work, so you can outperform the competitors in the market you choose. But this seems to be non-obvious to these teams and their leaders.

Researching the different symptoms and connecting the dots is what I focus on right now.

We already observed some. Increased pressure on development teams because the business goals aren’t met. Uncertainty how to improve business performance in the next quarter. Implementation of features for just one customer.

I’m pretty excited how the more complete picture will look like.

## Conclusions

We are really peeling layers of an onion here with our discovery work.

One approach would be to focus on the teams, who already want to implement discovery. We found some and are working with them. Even if they’re lacking a strategy, discovery helps them to ask the right question and establishing strategic thinking.

But we don’t double down on them because we intend to transform how our industry works.

My team is part of a software service provider. We are developing software agile for our clients and help them adopt an agile mindset. We want to serve them even better by help them to adapt the product mindset.

What is your experience in establishing strategic thinking? Where are you in your journey? Would love to grab a coffee.


I’ve talked to over 50 scrum teams in the past year about discovery.

Most of them weren’t ready to discuss starting with interviews, because the organization wasn’t managing with outcomes.

Most Scrum teams aren't ready for a discussion about product discovery


I’ve had a chat with a friend last week. He asked me how I evaluate whether an idea is viable or not.

“Isn’t it the innovator’s bias, that an idea will become profitable? So, how do you evaluate ideas in the beginning? I would love to found a start-up myself some time, but I have to become an industry expert first to make such a bet!”

To be honest, you don’t. At least not by planning. The most reliable way to find out is to learn along the way. Emphasis on learning!

But let’s start where you might be at now. Let me walk you through the way of developing a new product like a bet and fix this approach step by step.

## Starting with the MVP

So, you start out and know that it’s risky to build a new product. Usually, there are two ways people respond to that problem. They either want to work on the plan until they feel confident, or they attempt to make the first version as small as possible.

The first approach can become quite problematic. Launching a new product is innately complex. That means that you can’t plan what will happen in advance. Most plans crumble as soon as they meet the real world. There is useful information you can gather, but the amount is limited. Overdoing this step will make your plan more detailed, but not more reliable. This leads to false confidence in the plan. In the end, the founder gets so invested in her idea, that she isn’t able to pivot anymore.

The second approach is better. Only the user’s behaviour will tell whether your product will be successful. So, it makes sense to limit the investment in the first version to the bare minimum. This means that you will have much time, money, and energy left to adapt to the user's needs.

But what if nobody cares about the launch of your first small version? You lost the investment, and you can’t learn because there is no user to learn from. Luckily, there is a way out to know what the first version should look like without building it.

## Understand the customer’s problem first

How can we identify a problem worth solving without building a small product? Well, let’s think about what a successful product does. It is successful because someone pays for it, in some way or another. This means that it creates value, and it is able to deliver this value to the customer.

![The value stream of a successful product](/imgs/how-a-product-succeeds.png)

To be more specific, the value perceived by the customer has to be higher than the amount we want to charge for it.

Whether there is a problem with a high perceived value, is something you can research without building the product. How do people solve this problem today? Do they invest time or money? Are they frustrated?

At first, you can just research these questions. But you’ll reach a point where you can learn faster by conducting interviews.

If you have built up enough confidence or want to speed up the process, you can do a quantitative experiment. You can market your product without building it. This is called a Fake Door MVP. You build a landing page where you promote it. Then you drive customers to this site and push them to declare their buying intention. When they do, you have to put your cards on the table and say you haven't built this product yet. This feels rude, but it’s not what you say, it’s how you say it. You can thank them and offer a heads-up when you have launched.

This approach has a pitfall, though. What if you find a problem worth solving, but the market is too small? How many customers do you need anyway to be profitable? This is why I add a traction model to my modeling process.

## Model your business first to see what is truly critical

In our last approach, we’ve looked into the desirability of our product only. While examining that part, we should check if the market size might become a problem down the road. Or frankly speaking: Can we make enough money to be profitable selling it?

This equation consists of some variables. In its easiest form, it is the number of purchases multiplied by the price. But it is very hard to estimate how much you will sell. This is where business plans get very precarious. I often see assumptions of a market share you want to reach with your product. These are assumptions you can’t really examine early on.

A workaround is to start with a minimum success criteria. How much money do we have to make to succeed? This should be the bare minimum you really need.

Let’s look at an example. For simplicity, I’ll concentrate on a simple business where you sell a book. This is a one-sided business model with a one-time fee. But it works the same for more complex models as well.

We want to generate a nice little passive income of 120,000 € annual recurring revenue (ARR). So, let's set our price to 20 € per book. This means that we need to sell 6,000 copies per year to reach that revenue goal. Let’s break that down to a smaller time frame to test it faster. 6,000 copies a year means 500 per month or \~16,5 per day.

We want to sell this book on our website where we expect a conversion rate of 1%. That means, that 1% of our visitors buys one copy. This is just a guess, and we’ll adapt it while we learn down the road. With the conversion rate and the goal of purchases, we can calculate the number of visitors we need each day. 16,5 purchases x 1% conversion = 1,650 visitors each day.

- Is 20€ a price point the customer is willing to pay?
- When we want to sell this book on our website, can we meet the target of 1% conversion?
- Is it plausible that we can drive 1,650 potential customers per day to our website? (500 sold copies times 1% conversion rate = 50,000 per month)

Down the road, we might learn that one of our assumptions is wrong. Maybe it turns out, for example, that it’s just a book for a small niche. We did an estimation and 1,650 potential customers each day seems too much of a stretch. How you can do such an estimation is part of another article. In that estimation, we’ve assessed, that 825 visitors per day may seem more realistic. In that case, there are two other variables we have to adapt to meet our minimum success criteria. One possible lever is to double the price to 40 € per copy. The other is to increase the conversion rate to 2%. We have to decide what adjustments seem realistic and have to collect evidence about the new assumptions.

This is a real opportunity to learn fast and early. While this step seems complicated in the beginning, it should be a fast one. This is just the first model, and we have to learn through evidence, not thinking. As mentioned earlier, working too long on your business model can cause unjustifiable confidence.

## What does the plan look like now?

To wrap this up, how does the process look now?

1. Model a viable business and identify the riskiest assumptions.
2. Start learning by conducting _experiments_.
3. Adapt your business model step by step.

We built a model of different assumptions. They are neither true nor false. Our main job as entrepreneurs is to learn about them. And we should start with the most critical and the riskiest ones. If you find out that one is wrong, you have to adapt your model. This will lead to adjustments to other hypotheses as well.

You see that the business model is a very fluid thing, especially in the beginning. And it is a cycle that continues.

![The process of getting evidence for your business model](/imgs/how-to-get-evidence-for-your-business-model.png)

1. Create/Update your model
2. Identify the riskiest part
3. Gather evidence about that part
4. Update your model


## Conclusion

When you start out, you have a vision for your product. That particular version you imagine is probably not viable. But instead of being discouraged by that, you draft a viable model consisting of different moving parts. Each part is an assumption you can test separately. And every test creates learnings about your business model. This will force you to adapt your model until you find a viable one based on real customer needs. This one won’t be based on gut feeling, but on evidence.

The approach isn’t about science, though. It is about building confidence while moving fast. With every experiment, you attempt to gather evidence as fast and cheaply as possible. At some point, you’ll have gathered enough evidence for the moment of truth. Your next experiment will be actually the first version of your product.  🚀


I’ve had a chat with a friend last week. He asked me how I evaluate whether an idea is viable or not.

“Isn’t it the innovator’s bias, that an idea will become profitable? So, how do you evaluate ideas in the beginning? I would love to found a start-up myself some time, but I have to become an industry expert first to make such a bet!”

Get Enough Confidence to Build Your Product


When you get your very first users, your weeks are packed with research, refinement, and development. But you know getting feedback from your users is crucial to building a product that sells. You have to choose wisely which features to prioritize and have still to nail your value proposition.

So you start sending emails to your customers. But you get very little response. It feels like trying to shotgun blast and see who responds. So let me show you what customers motivate to give you feedback and how I ask them.

## What motivates people to give you feedback

There are two forces involved here. People's lives are packed with a lot of responsibilities. Plus there are so many things competing for their attention. This means your message should not just ask for an interview. For me, at last, it raises expectations of a quite boring meeting.

Another common piece of advice is to ask people for their opinion. This as well isn’t an exciting proposal in my opinion. Plus it remembers me of the many pitches of founders and proud product managers I've heard. They loved their product and craved for compliments and approval.

So what motivates them then? Most people with real problems love to talk about these problems. Especially if someone offers to solve them. So this should be your anchor.

## The actual Template

The template is inspired by the one from _The Mom Test_ by Rob Fitzpatrick to acquire prospects for problem interviews.

It consist of only two small paragraphs:

### 1. Who are you

You’re an entrepreneur/product manager/product owner trying

### 2. The problem they care about

You try to solve a horrible problem X, usher in wonderful vision Y, or fix stagnant industry Z.
Show weakness. Give them a chance to help you by mentioning the problem that you’re looking for answers on.
This will also clarify that you’re not a time-waster. Put them on a pedestal by showing how much they, in particular, can help.

## An Example

Hi Emilie,

I’m the founder of Acme Inc. the company behind the FindYourSoulmate app you are using. As you know we want to help you find your true soulmate and fix the evil dating industry.

But to be honest, I’m not happy with the onboarding experience of the app. We are losing many new potential soulmates there.

As you are one of our most active users, I would love to grab an online coffee with you. Let me ask you some questions about the onboarding and what did make you sign up for our app in the first place.

Love to hear from you

Fredi

## Conclusion

We have to be a little empathic with our users and it gets quite easy to acquire users for feedback interviews. Don’t waste their time and show them how they can invest their time valuable by stating the problem you try to solve.


You know getting feedback from your users is crucial to building a product that sells. So let me show you what customers motivate to give you feedback and how I ask them.

A template for acquiring feedback interviews for your product


Just as the title says.

I conducted 50 interviews without remembering anything really valuable.

So, I started thinking about the process and what was going on. In this article, I’ll describe our situation at first to give you a bit of tangible context. Then I’ll analyse the general problem behind our situation. Based on the to-dos you have in an interview, I'll show you how I failed on delivering on them due to our situation. Then we’ll dig into how you can prep for and conduct an interview when you search for a problem worth solving.

## Our situation
In my day job, I’m the Head of Innovation Consulting in an IT agency based in Bonn, Germany. We at tarent are excellent at solving complex problems with software on a large scale. But solving complex problems on its own doesn't make any money for our clients. We felt the urge to boost the value of the software by making sure we build the right things, before building them right. So, we decided to found a new business unit, which is called Innovation Consulting.

## Starting with a solution
From an entrepreneur's standpoint, we started with a solution: Startup methodologies. And we've had a group of potential customers: Different roles involved in developing digital products.
We didn’t expect any demand risk, as there are other innovation agencies who seem to do fine.
The problem with starting the solution is, that it can solve multiple problems:

- Faster Time to Market
- De-risk your idea
- How can I pitch my idea for a budget?
- Should we pivot or persevere?
- How to prioritise features by value
- What is the right value proposition to sell your product (irony isn’t it?)

Our goal was to determine which of the problems are worth solving for which customers. The next step on our journey was marketing our services to them. The challenge here is the combinatorics. The more problems you potentially solve and the more different customer segments you are serving, the more possibilities you are facing.

So, why is that a problem? It sounds like a good opportunity and more of a privilege. To understand that, let’s just have a look, what jobs are involved in conducting problem interviews.

## Your jobs in an interview
Problem interviews are used for discovery and the first validation with weak evidence. But to achieve that, you have to make sure that you are speaking to a potential customer. Only they can answer your questions.
Therefore, you have to qualify the interviewee first, before you really can dig into the problem domain. Only then it makes sense to investigate what they already tried to solve it to get a grasp of their commitment.
The schema is always the same:

1. Ask what their challenges are. Are they coming up with a problem you want to solve on their own?
2. If they don’t come up with a problem, zoom into a problem you intend to solve.
3. If you’ve zoomed in, make sure it’s really a challenge for them by asking what they tried to solve it. Are they a potential customer?

The goal is to conduct some interviews and get a grasp if there are enough customers, that have this problem and whether they care enough to pay you for solving it.

## So, what did I do wrong?
In my interviews, I tried not to prime them and asked about their challenges openly. The problem was, that they didn’t know what I was interested in, so they were talking about general challenges and how they solved them.
My problem was, that I wasn’t able to leave this altitude. I hadn’t thought through, how the different problems we could solve would manifest, in the customer's life. I must have missed so many non-obvious clues.  But to be really honest, until today, I don’t even know if they were potential customers. Where there non-obvious clues? I’ll never know. In the end, we’ve talked a little bit about their lives, but it didn’t help me to move our business forward.
## How do I prep for the interviews today
I have always a list of 3 top learning goals, like Rob Fitzpatrick describes it in his book _The Mom Test_. If I’m just starting out with a solution in mind, I model different scenarios beforehand. I aim for narrow problems. I sort the problems out, which aren't a perfect fit for my solution. Then I think about how the customer might experience the problems and how I can link this to the underlying problem. This prepares me to catch non-obvious clues.
This process already kicks out some problems. The rest gets prioritised by the team by anticipated value.
In the end, I try to find a topic for these challenges. In our case: Challenges starting a new project for a digital product like an app, website, or platform.
## How to navigate the problem space more quickly
When you start out with interviews, it can be overwhelming. Every so often I was so agitated, that I wasn’t able to capture all insights. And especially when you are just starting out, it can be challenging to zoom in one problem. There is always the concern to have zoomed into the wrong problem and miss the more important one.

A tool, which helps with that, is Customer Problem Stack Ranking.

With CPSR my interviews have the following structure:

1. I start with an open question: What is the biggest challenge starting a new project for a digital product?
2. I capture what they mention and dig into that.
3. I present my problems in no particular order and let them prioritise them.
4. I dig into the top problems

For me, this is the best of both worlds. I can check if this is a hair on fire problem by checking wether they come up with it on their own. After that, I'm able to challenge this by showing them the whole problem space. Did they miss an important thing? Didn't they think of it, because it didn't come to their mind, that it's solvable? Plus you have something documented you can compare easily.
One downside is, that this tool can make the interview quite formal. You can't just have an informal conversation. In my opinion, it's worth it, especially when you are just starting out or don't have much interview expertise.

A great tool for customer problem stack ranking is [https://www.opinionx.co/]


## Conclusion

Starting a business model from the solution is the hardest way to start a business. The more problems this solution can solve, the more are you drowning in different possibilities. The first thing you have to do in an interview is to qualify your interviewee. To be able to do that, you should have a clear understanding of the problem space you are currently trying to validate / invalidate. Start with an open question for current challenges, but be prepared to zoom in.


I conducted 50 interviews without remembering anything really valuable. What a waste of time. I’ll dig into what went wrong and how to prevent it.

I conducted 50 customer interviews without any insights


When I talk to founders and product people, I see packed Lean Canvases all the time. When I ask how much time went into it, they say proudly weeks, even months. But also, that they aren’t really satisfied with it and have the feeling, that “it’s not a good plan yet”. They want more focus, it’s not coherent enough or not clear and precise enough. They aren’t ready to move on and don’t have the confidence in this business model to continue with the process.

What a pity! The fastest path to a neat Lean Canvas is to start.

In this article, you’ll learn what the actual purpose of a Lean Canvas is. So you'll know what you need on it to move on, so you could take the next steps.

## The purpose of a Lean Canvas

To know when your Lean Canvas is ready, you have to know what it is for. An obvious feature of a Lean Canvas is to neatly and concisely communicate your business model on one page. And that is the reason, why you don’t have the feeling of being finished. It’s so easy! You simply don’t start with a neat business model. It’s a vague guess. You have to develop it. So when you start out, the Lean Canvas has two purposes:

Identifying the riskiest assumptions in your business model

Checking for plausibility

So have a look at your Lean Canvas. And if you don’t have one at the moment, pause here and give yourself 20 minutes to draft one.

I bet it’s messy.

There might be blank fields.

I bet you’re feeling a little bit embarrassed, that this is your current state. And don’t feel bad if you have already been working on your start-up for some while. I’ve worked for two years for a start-up. When I look at the business model today, I can’t believe how little progress we’ve made. We were so busy developing software instead of the business model.

That’s ok. Let’s change it. But not by thinking, but by gathering data! You are going to learn so much and so much faster than by thought experiments.

## How much time should you take to fill the Lean Canvas?

20 minutes! Why so little you might think. Because we are entrepreneurs and not scientists. We can’t afford to spend 3 years thinking about our business model. The world is moving too fast and we don’t want to wait 3 years until we try being successful.

So the fastest way to make progress and tangible results is by moving on and gathering data to iterate on your business model. Your lean canvas will become more beautiful and complete in each iteration.

## What is the minimum goal?

We’ve identified, that you want to do some experiments to develop your business model / Lean Canvas. What is the prerequisite of doing so? Every successful business model is based on creating, delivering, and capturing value. That means, that you have to create something that is so useful to somebody, that they are willing to pay for it.

The first goal in business model development is always to identify a problem worth solving. So people are willing to pay for your solution. The bare minimum needed to do that is to know who you want to do research about. And that is a quite tricky question. Because you want to identify a customer segment, which has a uniform problem. This is often not the case for a group defined solely by demographics like age or gender. Instead, you want to find some psychographic. Psychographics are characteristics like subconscious or conscious beliefs, motivations, and priorities.

If you don’t know any psychographics do a thought experiment. First brainstorm 2-3 problems you want to dig deeper into. Then try to identify a subgroup by asking the following questions:

- Within this group, which type of person has this problem the most?
- Is everyone within this group really trying to solve this problem?
- Why does this sub-set have this problem more? Is there another more specific problem?
- What is the motivation to solve this problem? Does everybody in the whole group has this motivation? Why not?
- What additional motivations are there?

If you want to do quantitative experiments like prototypes (aka Fake Door MVP), you have to have a specific solution and value proposition as well. Check if they really fit with the problems. Do a marketing one-liner for a sanity check:

```
Have you had <PROBLEM>? I do <MY SOLUTION> to solve it.
This results in <SUCCESS THE CUSTOMER WANTS>.
```

Check if the success is really linked to the problem.

## Conclusion

To sum it up real quick: You should invest 20-60 minutes max in the first version of your Lean Canvas. This version of the Lean Canvas is just a starting point for conducting experiments. To do them efficiently you need to know how you reach your customers. So the customer segment has to be specific enough for that. Additionally, you should have an idea of 2-3 top problems you want to investigate.

If you aren’t familiar with this customer segment, don’t worry: It’s just a starting point and you’ll get there. If you worry about missing a much more viable problem, remember that you want to act fast and don’t want to end up in analysis paralysis. If you can't get rid of the feeling, that you are missing a more valuable problem, do the following. Draft multiple narrow Lean Canvases with all problems you could think of. Focus on the fields customers, problem and solution. Then you could prioritize these Lean Canvases and start with the first one.

__Get my next article by subscribing to my mailing list: Define your first experiment for finding the customers to your solution.__


🤓 How to get  your business model from a vague messy thing to a neat, narrow and focused model …
✋ It’s not by fiddling about it.
🤌 I get it, I’ve done it a hundred times!
🚀 The irony is, that the fastest way to get it nice and narrow is to start.

Business Modelling

Finish your Lean Canvas already


Most entrepreneurs I talk to endorse that following the build, measure, learn cycle is a good idea. So a fundamental part is measuring to learn from the data. Do you want to answer questions like how many customers are visiting my landing page? Where do they come from? How many of them convert? How many of them even scroll down to the Call to Action? You can and should leverage multiple analytics solutions for that. E.g. you could augment Google Analytics with Hotjar, to get more qualitative insights about how customers behave on your website. To easily install and configure these solutions you can use Google Tag Manager. It even helps you define custom events you want to track without coding. So let's dive into it.

To kick things of I'll walk you through the setup of Google Tag Manager (GTM) itself. Next we‘ll create  a Google Analytics 4 *property.* A *property* represents A place, where you want to collect data. Then I‘ll show you how to integrate it in your website via GTM. The last step will be testing of the GA4 configuration and deployment to production. You‘ll end up with basic tracking of page views, link clicks and scrolls (scrolled down to the 90% mark of a page) and have the basis for installing other marketing tags or track custom events.

Consent mode is also a pretty cool new feature, but covering that in this post would be too elaborate. I'll save it for the next post, so make sure you're subscribed!

Disclaimer: In this article, I'll only talk about Google Analytics 4. This is the new version of Google Analytics, launched in late 2020. The difference between Google Analytics 4 and the *old* Google Analytics Universal is covered on [this page](https://support.google.com/analytics/answer/9964640?hl=en#zippy=%2Cthemen-in-diesem-artikel%2Cin-this-article).

Another Disclaimer: I can't provide you with any legal advice. You are responsible for your implementation and for being legally compliant. Consult your data security officer and lawyer for a final check.

## What is Google Tag Manager?

Why should I bother, to set up Google Analytics via Google Tag Manager? And what is Google Tag Manager anyway? This is easy to answer:

1. Manage the integrations of different analytics and marketing solutions (aka Marketing Tags)
2. Track custom events like form Submissions or that users viewed an element on the website
3. Leverage the new consent mode to get more insights in the future, while being GDPR compliant.

Google Tag Manager is a tool to set up marketing tags on your website without the need of writing code. So you might ask yourself: What the heck is a marketing tag? At least I did the first couple of times. It is a code snippet (with the name tag like in HTML tag) to set up marketing tools on your website. This can be classic analytics solutions like Google Analytics, more qualitative analytics solutions like Hotjar for Heatmaps or advertisement platforms like Google Adwords.

You have to integrate the Google Tag Manager on every webpage of your site or page in a mobile app. This enables you to manage at what point different marketing tags should be triggered on your website - without leaving GTM.

## How to set up Google Tag Manager

This part consists of three steps: Createing an account in GTM, integrating it in your website and testing your setup. I describe the first two steps very briefly. The Tagm Manager UI guides you through the creation of the account. The technical integration highly depends on the technologies used to built your website.

### Create an GTM account

1. Go to tagmanager.google.com
2. Create an Account with:
    1. A descriptive account Name like: Acme Inc.
    2. Your country configured
    3. And the configuration for your first GTM container
        1. The URL of your website like [ga-gtm-talk.netlify.app](https://ga-gtm-talk.netlify.app/)
        2. The target platform, where you want to use this container

![Google Analytics 4 Account Creation Screen](/imgs/google-analytics-4-new-account.png)

### Technical Integration

Install the GTM container on you website. ([https://developers.google.com/tag-platform/tag-manager/web](https://developers.google.com/tag-platform/tag-manager/web)). This step depends on the technical implementation of your website. You might want to talk to your developer.

```jsx
import Document, { Html, Head, Main, NextScript } from 'next/document'

const GOOGLE_TAG_MANAGER_ID = process.env.GOOGLE_TAG_MANAGER_ID;

export default class MyDocument extends Document {
  render() {
    return (
      <Html>
        <Head>
          <script dangerouslySetInnerHTML={{
            __html: `
              (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
              new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
              j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
              'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
              })(window,document,'script','dataLayer','${GOOGLE_TAG_MANAGER_ID}');
            `
          }}/>
          <link rel="stylesheet"
            href="https://use.fontawesome.com/releases/v5.11.2/css/all.css" />
        </Head>
        <body className="antialiased bg-body text-body font-body">
          {/* Google Tag Manager (noscript) */}
          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=${GOOGLE_TAG_MANAGER_ID}"
          height="0" width="0" style={{display: 'none', visibility: 'hidden'}}></iframe></noscript>
          {/* End Google Tag Manager (noscript) */}
          <Main />
          <NextScript />
        </body>
      </Html>
    )
  }
}
```

### Test your setup

After deploying the code changes, you are ready to test your Google Tag Manger Setup. There is a Preview button in the upper right corner. Click it and the Tag Assistant opens for the current version of the Workspace.

![Google Tag Manager UI Header](/imgs/google-tag-manager-ui-header.png)

After entering the URL of your web page, you'll see the tag assistant. It shows in the header that one GTM Container was found. In the body of the page it says that no tags have been evaluated, because, we didn't configure any.

![Google Tag Assistant with empt GTM container](/imgs/google-tag-assistant-empty-container.png)

## Setting up Google Analytics with GTM

For the setup of your analytics, you have to create a GA4 property. With the measurement ID of that property you can configure the GA4 configuration tag in GTM.

Start with creating a new Google Analytics account, if you don't have one. Follow the wizard, in which you'll have to provide some personal information and confirm your email address and phone number.

To create a new property in Google Analytics, you have to open the admin section by clicking on the gear icon in the bottom left corner.

![Google Analytics 4 New Property](/imgs/google-analytics-4-new-property.png)

![Google Analytics 4 Set Up New Data Stream](/imgs/google-analytics-4-set-up-data-stream.png)

Then click the *create* *Property* button and follow the wizard. After successfully creating the property, you need to create the data stream, which is used by GA to collect the events. For your website choose "Web", enter the URL and give the stream a name. Let *enhanced measurements* selected, to have some basic insights already set up for you. On the next screen the Measurement ID is displayed in the upper right corner.

Now you can configure Google Tag Manager to deploy Google Analytics to your website. Go to [https://tagmanager.google.com/](https://tagmanager.google.com/). We want to create a new tag, so click on Tags and hit the New button. A Tag consists of two sections.

The tag itself, where you configure what data is collected and to which platform it is sent. In the second section, you can configure, when this tag is going to be triggered. You want the basic analytics tags on your website to be triggered on every page. More specific tags might need something like a trigger on an outbound link click.

Click on tag configuration and choose the Google Analytics: GA4 Configuration Tag. This tag configures your website for Google Analytics tracking. The basic setup includes tracking of basic events like page views. The property setting enhanced measurements adds other good insights like link clicks to it. Now enter the Measurement ID of the GA4 Data stream into the appropriate field.

Next click on the triggering section and choose the pre defined trigger "All Pages". After naming this tag in the first line you can hit save. The tag you've created is now a draft in your workspace. Next, I'll show you how to test and deploy it.

## Test your Setup with the Tagassistant

As described in the first section, Google Tag Manager enables you to test your newly configured tags without deploying them to your production website. To do that, click the Preview button in the upper right corner of the GTM UI. This opens the Tag Assistant for the current version of your workspace and asks you which URL of your website you want to view to test your tags. Hitting Connect leads you to a screen where you can see what Google Tag Manager did on your website. This refreshes when you interact with your website in the opened window.

On the left you see the messages that have been sent. A message is new data in the GTM data layer. You can think of it as the way developers can interact with GTM. If you click on a message, you can see the state of GTM after evaluating this message. You can see which Tags were fired, the value of the different variables, the state of the data layer and errors that occured while evaluating the tags.

Opening your website via the tag assistant sets your Google Analytics containers into debug_mode. That means, that you can see the events in the DebugView in Google Analytics as well.

### Deploy it

After we've ensured, that everything is fine with our changes we can finally deploy them to our live production page. You do that by hitting the send button in Google Tag Manager in the upper right corner of your workspace.
You can give the new Version a name and description and review the changes you are about to deploy. After hitting send you are done and your changes are in production.

## Summary

In this article, you've set up Google Tag Manager and Google Analytics 4. Now you know how to leverage Google Tag Manager to integrate marketing tags. A plus is that your Google Analytics collects data for valuable insights already. Your setup isn't GDPR compliant though. You are not asking for consent before you start tracking your customers with the help of cookies. Another key feature you don't use is the ability to create custom events and custom triggers. You can use them to collect more data without the need to code.

The consent mode and setup of custom events with custom triggers will be part of future articles in my blog. If you don't want to miss them, feel free to subscribe to my mailing list.


Most entrepreneurs I talk to endorse that following the build, measure, learn cycle is a good idea. So a fundamental part is measuring to learn from the data. Do you want to answer questions like how many customers are visiting my landing page? Where do they come from? How many of them convert? How many of them even scroll down to the Call to Action? You can and should leverage multiple analytics solutions for that. To easily install and configure these solutions you can use Google Tag Manager. It even helps you define custom events you want to track without coding. So let's dive into it.

Entrepreneurs Analytics

Integrate all your Analytics and Ad-Tracking with Google Tag Manager

Can I already charge for my new product? "Never, I'm too ashamed of the current state!" - I get it! I've been there. But there are important rationale, why it's bad to give your MVP away for free.

The right price for your product with these 3 questions

Every Software Engineer loves working from home and the distrusful management hates it? It's different to us. With mob programming it's even possible to build a new team of software engineers easily.

Remote Team Building with the Help of Mob Programming

Automated tests guarantee ongoing productivity. This article leads you through adjustments to the system under test, debugging them, setups and teardowns.

Setup and Teardown of Automated Tests

Dino Omanovic and I compared the programming language Go to others in different categories. From memory consumption over robustness and deployments to steepness of the learning curves. We've covered it all and gave you some tangible insights to figure out, wether Go is a match for your next project or not.

Build a productthat sells

Digital Entrepreneur Insights

Empower your product team with discoveryand strategic thinking bottom-up.