Steve Sanderson🚢

/stevesanderson

Steve Sanderson🚢

I love making important things better, while glimpsing at the patterns that connect

User Avatar

Steve Sanderson🚢

4y ago

The Art of Exaggerating - Overshooting To Find The Target
Steve Sanderson @stevesanderson

You probably already know how important assumptions are when navigating any landscape of uncertainty, so you also know it's important to elicit those assumptions. One important step is to increase the precision of those assumptions.

After doing this on Product, Engineering and other teams decades, one of my favorite language patterns to increase precision is to use a bit of exaggeration:

challenge and find the limits of an assumption by exaggerating

When you use exaggeration to restate their claim, you are restating it by "overshooting" it beyond what they likely meant, and typically people will correct it to be more precise.

It goes something like this

Someone states their assumption like

"teachers have to be skilled in X for students to be able to learn X"

Your challenge sounds something like

"students can never learn X without teachers already knowing X"

Then step back and let them respond. Most likely by the end of the discussion, you'll end up with a more nuanced assumption, such as

"teachers who are skilled in X increase the odds students will learn X"

This works, in part, because of the rapport you have with your colleagues let's them correct you to be more precise.

And that leads to one other tip:

make sure you already have rapport before challenging with exaggeration, and that you maintain rapport the whole time

If there's insufficient rapport, especially when there's a power differential (e.g manager vs managed), it'll be harder to get accurate results. Put another way, don't be a jerk - it's unnecessary, and will distort your results anyway.

Using this language pattern, in the context of rapport, is one of the most effective ways to understand the boundaries of what other people think. And that's a good thing.

User Avatar

Steve Sanderson🚢

4y ago

When It Comes To Change, Indifference Is Your Enemy
Steve Sanderson @stevesanderson

Products introduce change for people, hopefully to create value for them and the business. When you're attempting this change (whether for a person or a system of people) change,

indifference is your enemy

I've been working in and around Product for over 30 years now, and this one pattern changed how I approach my work.

If your change triggers an emotional response (like anger, sadness, joy, etc.) then you're probably on the right track.

When I was working on Food on the Table with @manuelrosso, we interviewed groups of friends to learn more about their problems with regular ("Tuesday Night") grocery buying, cooking, dinnertime.

These were a series of interviews we held in someones beautiful home, and was casually set with some wine & cheese and had several people around a dinner table.

To my surprise, more than once the conversations resulted in someone having strong emotions: crying, laughing, angry, etc.

While I understood how they'd gone from talking about serving dinner to grief about being rejected by a step-child, I didn't think those strong feelings were a good sign.

During one of those times, Manuel explained it to me:

we want to identify problems with strong feelings, it's indifference that's the enemy

One other tip helped when strong feelings were in the room:

don't judge their feelings with your beliefs

Recognize the reality of their feelings, and consider being grateful they're not indifferent!

h/t @manuelrosso for introducing me to this idea

User Avatar

Steve Sanderson🚢

4y ago

You preparing to build a new feature ... and you don't want to waste time and money with no impact
Steve Sanderson @stevesanderson

Context

You've got a new feature you're preparing to build. You're convinced there's a problem worth solving.

Problem

You're uncertain about the demand, and you don't want to waste time and money building something that won't have an impact

Solution

Offer the 1st step of your solution and put them on a waiting list

  1. Rough design the user-facing aspects of the feature

  2. Implement the first point of interaction with the user

    • log when a user views the where the first point of interaction occurs

    • log when a user interacts with the first point of interaction

  3. Implement a response to their interaction by notifying the user that this feature is coming soon

If you want to know more about their needs, consider

  • prompting them to describe their needs

If you're concerned about the consequences of not delivering the benefit, consider

  • revising the language to shift away from offering the feature and towards telling them about the possible feature, e.g. instead of "get report on usage" revise it to say "find out more about usage reporting"

  • prompting them to let you know if the need is urgent, and then fullfill their need with a concierge MVP

If you're concerned about the whether they're window shopping or actually have a need, consider

  • prompting them to actively choose to tell you they're interested, e.g. "yes I'm interested" and "no thank you" buttons

capabilities you'll need

  • to decide ahead of time what success looks like, e.g. what % of people who view the first point of interaction eventually interact?

  • to view and analyze the event stream (i.e. the views, the interactions, any optional entries, responses, etc.)

  • to immediately turn off the experiment

  • to serve any concierge MVP work that comes out of the experiment

management systems you'll need

  • team monitors and adapts/decides based on the results of the event stream

  • terminates the experiment when done

  • track and deliver any concierge MVP work

Caution - even the narrowest experiment may test other variables

  • in this case, if you don't have a proven, predictable, method for customers to discover a new feature then you are also testing how customers become aware of a new feature in your product.

User Avatar

Steve Sanderson🚢

4y ago

reader beware: this is basically an opinion, even gut-feel, piece - backed up with personal experience and inspirational stories.

You're in a successful product company, and successful companies, with successful products, typically want to build new successful products.

In successful companies, you (hopefully) have a lot of data about customers & users. With that, you can find large opportunities, and jump straight to quickly building the features that best serve the largest number of that group.

However, there's likely one factor you completely missed - the path starts with innovating at the margins

Innovation is often path-dependent, i.e. it often works better to go from the margins to the mass market. Starting with targeting the solution to the mass-market (i.e. serving the "largest" group of people) is often a flawed path.

In startups, this is never a choice: most of them have to start at the margins, whether solving their own problems or solving problems for a small group of eager, early adopters. In my personal experience, this has been consistently true: a repeated pattern is that we started narrow before we went broad.

In disability tech, they've chosen the margins as their focus, and also follow the repeated pattern of starting narrow before going broad, e.g.

Solving problems for people in the margins not only lets you get early traction, and therefore more rapid learning, it appears to make the solutions more resilient - i.e. if they work for the people in the margins, then it's more likely to work for the mass-market.

Starting with the mass market and moving to the margins appears to be less effective, plus there's often pressure to move on to a new large opportunity instead of targeting the margins.

Follow a strategic path of innovation, from narrow to broad

In your product effort within a successful firm, identify the large opportunity, and start first with a segment of outliers.

Consider,

  • When you hear "this will work well for the vast majority of those users" flip the conversation and ask "what amount of this group are we willing to leave out?"; consider if this is an outlier segment to start with, and the impact if you left them out of the success.

  • When there's a demand for efficiency - reorient to make sure efficacy has come first.

  • When the objection to starting narrow is an assumed delay to delivery, remind yourself that in a context of uncertainty, speed of learning is a strong predictor of success.

Teams that attempt to jump past learning from the margins are making an expensive gamble that they will get it right on the first try.

Instead, start narrow (prove efficacy, build resilience), then go broad (prove efficiency and scale).

User Avatar

Steve Sanderson🚢

4y ago

Fully immersed in edtech since 2016, with my venture in 2005, I've only seen companies that have been governed in some form of "shareholder capitalism", i.e. maximize the shareholder value. But what if there's another form of governance that yields a higher total impact?

🧵👇

More recently, companies have expanded, or shifted, their governance to consider other stakeholders, i.e. students, parents, community, etc. - witness the rise of B-corps as an example.

In parallel, there are venture funds with an intentional consideration of those stakeholders, expanding "shareholder capitalism" to account for the impact on stakeholders, e.g. @reachfund, @sjfventures

However, I've yet to see any edtech ventures that go as far as ownership down at the stakeholder level, i.e. an edtech venture aligned to something similar to what @zebras_unite advocate.

Imagine an edtech venture that not only delivers impact to students, parents, teachers, and administrators, but also where they are owners, i.e. like a co-op, and therefore they accrue benefits, and influence governance.

Through the lens of my roles as CEO, CPO, or CTO, I believe it's very possible that total impact when governed in some form of "stakeholder capitalism" would be noticeably greater compared to "shareholder capitalism".

However, I also believe the total financial results would be greater when governed in some form of "stakeholder capitalism."

This leads me back to my question: can anyone point to an edtech venture built on some strong form of "shareholder capitalism" (such as a co-op)? Are there any in progress, successes, or failures?

User Avatar

Steve Sanderson🚢

4y ago

In the previous post, I introduced the idea of using the perspective of supply and demand to begin making sense of how a product team operates. In this post, I'm going to give some examples and the practices I use to get oriented on their work.

🧵👇

With a supply-side team, I might encounter something like:

  • that the large majority of their time was spent focused on what they could build, and little to no time was spent on learnings from users

  • stories that motivate them such as "we shipped this new feature, yeah!",

With a demand-side team, I might encounter something like:

  • that they spend time learning from users which affects decisions in the time spent building

  • motivating stories such as "we found this really painful (customer) problem"

To get oriented, I go through two big steps:

  • simplify my work with a few assumptions

  • ask a couple of questions

Assumption 1: it's more useful to learn what's actually true, not what people claim is true.

Assumption 2: the purpose of a product effort is to impact one or more groups of people (i.e. impact on customers, users, business, etc.). I don't assume which group, or whether they're even customers.

Assumption 3: shortcut by categorizing practices in product work as "supply-side" (e.g. writing code) or "demand-side" (e.g. testing customer problem hypothesis); though never absolute, i.e. context matters (e.g. "write code to test a customer problem hypothesis" is demand-side).

Get an idea about their attention direction (is the team focused internally or externally?), by asking "how do you define success?" (or "how will you know when you're done?");

is the focus of their answer:

  • external - is it about some users outside of Product making some form of progress (e.g. users happier, salespeople more effective, etc?)?

  • internal - is it about the Product group's own work (e.g. delivered feature, beautiful design)?

Where an external focus implies a more "demand-side" stance, and an internal focus implies a more "supply-side" stance.

Get an idea about their relationship with the users, by asking "what happens when the users fail to make progress with the product" (or "what happens when users can't do their work");

do they see themselves:

  • relational - concerned with, and interested in learning more, curious, or even consider themselves (partially) accountable for helping users?

  • transactional - separate (e.g. siloed) from users: "the product is functional, their problems are their problems, and not ours"?

Where relational implies a more "demand-side" stance, and transactional implies a more "supply-side" stance

Understanding whether a team tends more toward "supply-side" vs "demand-side" is a big part of assessing whether the team is a fit for their environment. There is no absolute best stance, it depends on the environment.

Simplistically, a demand-side stance is better in an environment of uncertainty and a supply-side stance is better in an environment of certainty.

Of course, once you can see the (mis)fit, then there's the question: if the fit isn't right, how can you help to improve the fit? I'll save this question for another post.

h/t to @chrisvmcd for his video on Social Practice Theory

User Avatar

Steve Sanderson🚢

4y ago

As an unapologetic generalist, I've been actively engaged in using experiments to learn across many domains, levels and industries, since 2009.

And yes, it happens at home as well...

🧵👇

Over the years, across the different contexts, up and down the org. chart, I have learned that most people who want to become skilled at learning through experimentation end up failing because of fear: that the risks will outweigh the benefits.

When I was first challenged to experiment ("on pricing"), I was fortunate enough to be challenged by someone who found the right argument to help me shift my thinking ("you don't have enough of a brand to damage"). Brutal, but effective.

Unpacking that fear, there are a number of causes behind it, including:

overconfidence in the answer - they're certain they're correct & there's no need to learn more; they can't or wont recognize they're operating in an environment of uncertainty

don't want to know - making anomalies visible might be challenging:

• personally - ego investment in being right; "I don't want to be wrong"

• culturally - no accommodation for adapting plans in response to anomalies; oftentimes discouraged with middle-management

too expensive - they don't know it can be cheap and fast; they overbuild the experiment; it's natural early on, & the best remedy is

• a context where the cycle time is very short, combined with

• skills at minimizing the size of experiments (cut in half, then in half again)

big-bang - they don't see a way to "ease into" becoming skilled at the practice

One way or another, the learnings happen - the only real question is will you learn quickly & cheaply (by experimenting) or take a long time and spend a lot of money (by building the whole thing)?

User Avatar

Steve Sanderson🚢

4y ago

The Tip I Would Give Someone Who Wanted To Get Started Working In "Product"
Steve Sanderson @stevesanderson

Do you want to get started working in Product?

I have been working in and around Product for over 30 years now.

While the practices around Product (product management, product led X, product thinking, product development, product design, etc.) have emerged over time, there are several strengths necessary to be great at Product.

Three of my favorite are:

  • empathy - for those you serve; not just users, and customers, but salespeople, installers, support, regulators, etc.

  • a bias towards action - combined with empathy, a bias towards action yields compassion; you need to act to make something better for others

  • juggle multiple systemic points of view - you can understand a system in which people are embedded, you can hold multiple systems in your head (despite conflicts), and can work to resolve them (especially where they conflict) to deliver progress to those you serve.

So, one tip if you want to pursue a similar career path is:

Start from customer support/customer success and help solve the right problems

Some of the most talented Product people I've worked with came originally from customer support/customer success. They came to Product already skilled with one (or more) groups (customers, users), and with compassion. With their exposure to more groups, e.g. the needs of the business, the needs of salespeople, etc. they built up the muscles to juggle the multiple systemic points of view.

More generally, if you are not in a customer support/customer success role, then find a role, or invite yourself into a role, where compassion for two or more groups is needed and strengthen your muscles by solving the conflicts.

Depending on how silo'ed your organization is, you may meet some resistance: "stay in your lane". Take nothing for granted, are there opportunities to get more involved outside your "lane"?

User Avatar

Steve Sanderson🚢

4y ago

That time I was CEO with a fatally long sales cycle, and "dumpster diving" drove the pivots that fixed it

🧵👇

I was the CEO for EdSight, and while early on we had evidence we were on the right path, when we realized that the pieces were not actually fitting together, two unexpected pivots led the way out.

To understand, let me back up and give some history.

Before EdSight, I was the CEO for Double Line, a professional services company with a great history changing the way data management K-12.

When I joined Double Line as the CEO, one of the goals was to create and execute a product strategy driven by their successes in services.

With the insights from the experienced team and customer discovery with our various clients and partners, we were able to test out various ideas.

Eventually, we found what appeared to be compelling demand

With enough evidence, we spun a general-purpose toolset that helped K-12 educators improve on their quantitative outcomes out into a new company, EdSight, and continued.

We found problem/solution fit and some early product/market fit.

Eventually, however, cash flow dependencies collided with faulty assumptions about the sales cycle.

In short, we were not going to survive the long sales cycle.

As a team, we looked at options but kept coming back to something we'd often heard during sales: "love the product, love the idea, and once we put out this dumpster fire, we'd love to buy it."

Up to then, we'd always tried to change their minds.

This time, we looked in the dumpsters.

What we discovered changed everything.

They were telling us about some problem that was similar, but not the same, to what we were solving.

They were small, immediate, high-stakes data-related problems that had to be solved.

1️⃣Our first pivot was their problem & our solution

When we dove into those problems (no pun intended), we picked one problem we could solve using a subset of our existing product with only 1 new feature.

When we went to market offering to solve their narrow problem, with the same product + 1 new feature (immediately implemented via concierge MVP).

We found immediate, surprising, traction.

2️⃣Our second pivot was our pricing model.

We found out about how some buyers have funds available immediately if it falls below their discretionary budget.

We took the new offering with new pricing to market, and the sales conversation was dramatically different

the sales calls became about closing, sometimes even before we could get through the our pitch.

✅We pivoted from a high-dollar, long (12 - 18 month) sales cycle to a low-dollar, short (30 - 60 days) sales cycle and started closing deals immediately.

We refined the messaging to the point where cold emails to K-12 administrators would predictably result in 1st and 2nd sales calls, sales calls became about closing (sometimes even before we could get through our pitch), and closes that averaged ~60 days.

As a result, our cash flow needs and the sales cycle no longer conflicted and we were on the right trajectory.

Further, we shifted our strategy to build out a series of niche solutions, on top of a more general platform, using their adoption to "pull in" the necessary infrastructure that enabled us to solve more problems.

Sadly, unrelated to the above: the niche problem we solved was mandated by a State education agency, and when the pandemic hit - the mandate was suspended for two years. Unsurprisingly, demand evaporated immediately.

🤔A few takeaways for me:

even classically long sales cycle deals (K-12 administrators!) may have a way of becoming short

being able to turn the knobs on all aspects of the business model increases the odds of "fit"

macro-changes are brutal

🙏h/t @PolicyRiot for the tip on the discretionary budget in K-12

User Avatar

Steve Sanderson🚢

4y ago

I've mentored, advised, coached 100's of product teams in startups & larger firms.

want to know a secret?

there is almost always a way to test your product ideas manually (aka concierge MVP)

The five strategic moves I use and recommend, contrived story to illustrate

🧵👇

1/5) get clear on the people and their problem

Starting with the problem? Then make sure you can connect with those people with the problem.

Starting with a"solution looking for a problem"? Spend the time to uncover the problem you're solving, then connect with the people.

2/5) get clear on their benefit

Do they "roll out of bed in the morning" wanting the results of your product?

If not, keep working up the value chain, i.e. ask repeatedly "what will it do for you if this is solved?" until you find that benefit.

3/5) Try the "as-if frame" to uncover how was this solved before / elsewhere.

Ask yourself: in the past, before technology, if someone had this problem and wanted this benefit, how would they have accomplished it?

4/5) Now, implement the concierge MVP by subsidizing (i.e. your time, your money) solving the problem.

Use how it was done before / elsewhere, while offering it with your new business model (i.e. revenue model, performance, etc.).

This is not supposed to be directly scalable.

5/5) Successful delivering the benefit & collecting revenue?

✅ you've converted an "uncertainty problem" into an "optimization problem", now convert the expensive, manual work into scalable, automated work.

❌ keep iterating until you're successful or you pivot.

contrived story

Imagine X, and team, have a brilliant solution to help home gardeners maintain beautiful landscapes, a "machine learning driven garden monitor that will detect needs of garden plants and dispatch crews to take of the plants as needed"

They assemble a machine learning expert, computer vision expert, product developers, product designers, etc and are ready to roll.

Before they pull the trigger on building, they want to check on the validity of their problem/solution fit.

The problem is "home gardeners want the certainty of beauty, but with only some of the work"

The benefit is "their gardens will always be beautiful, regardless whether they do the work or not"

The "as-if frame" yields "in the past, dedicated gardeners (with lots of $)"

Implement by subsidizing some form of the past solution (e.g. X & team spend their own time and $ to monitor gardens daily and dispatch gardeners),

through existing customer channels (e.g. text), with a revenue structure of their new business (e.g. $10 month + hourly for crews)

As the concierge MVP, it's not scalable (e.g. multiple adults doing manual work, etc.)

If it's successful, optimize the biggest obstacle (ML? Cameras? Order taking? etc.)

If it's not, iterate and work on the biggest / riskiest "wrong"

h/t @ericries , @manuelrosso

User Avatar

Steve Sanderson🚢

4y ago

Is there a shift coming in EdTech data management?
Steve Sanderson @stevesanderson

We now live in a world with higher awareness and expectations around data privacy and ownership.

The awareness of the uses, and misuses, of data has increased, and is strongly correlated with SaaS (centralized data).

EdTech is right there, with the current successes typically SaaS-based, with centralized data managed by the vendor.

Up to now, vendors address this with a combination of regulatory and technical pieces, but almost always assuming centralized vendor management:

  • regulatory - voluntary and mandatory standards for vendors

  • technical - architectures and implementation to protect centralized data

I believe the demand is increasing for progress on EdTech data management, e.g.

  • increasing the the distribution options (i.e. where the data resides),

  • increasing the control by the owner (i.e. more fine-grain control of data access), and

  • minimizing the gap between who the data is about and who owns it (i.e. creators of data actually own it)

If we apply this to EdTech, could we end up with a future where student data is managed like this?

  • distribution - distributed per the individual owner choice (instead of centralized by vendor)

  • control - where each owner has fine grain access control (instead of controlled by vendor)

  • ownership - where the student / parent / guardian directly owns the data (instead of owned by the school system)

User Avatar

Steve Sanderson🚢

4y ago

What if your customers helped you with discovery?
Steve Sanderson @stevesanderson

Imagine you're a busy product manager of a successful product, and you're thinking about what's down the road... and you want to identify new opportunities.

You could focus on known, existing problems, and solve them with new features in your existing product... or you could try expanding your impact or even expanding your market.

The problem is, how can you explore this space quickly, and with low risk?

This pattern starts with the assumption that there are certain features in your current product that may be a sign of hidden demand.

Start by finding those features where users may be doing more with your results than you can see, e.g.

  • any features (now or in demand) that generate Excel or CSV files?

  • any features (now or in demand) for reports or dashboards?

Follow up by finding out what actions your users (want to) take with those results, e.g.

  • Excel / CSV - what actions do they take with the data?

  • Reports / dashboards - what actions do they take with the dashboard or report results?

Then get to deeper meaning of their actions, e.g.

  1. pick those actions that include

    • any further processing or analysis, and/or

    • communicating the results to someone else

  2. and ask your users some form of "what problem(s) do these actions solve, and who they are solving them for?"

Now you know there's a problem, that they're aware of having a problem, and that they've put together a solution out of pieces parts.

Congratulations, you've addressed 3 of the 5 levels of Steve Blanks' earlyvangelist Pyramid.

User Avatar

Steve Sanderson🚢

4y ago

Don't Hire the "BEST" Developers
Steve Sanderson @stevesanderson

If you are part of a group solving some problem you've never solved before, especially if you're not a technical leader, then odds are you've got lots of well-meaning advice about getting the "BEST", most cliche-superlative-filled technical folks.

Regardless of what "technical" is, the advice goes, hire the "best" (software developers / designers / biologists / doctors / etc) you can get.

They wouldn't be wrong, but the problem is - they are not right.

They are not right because of what they haven't told you: context matters & what's best for your situation isn't likely what was best for theirs

At an early stage - let's call it "exploration" - your job is to discover, and hopefully validate, the right demand: demand to solve a problem, demand that causes them to pay in some form, demand that has a large group of people, demand that you can fulfill, etc.

To explore well, you and your group must be highly adaptable, i.e.

  • one day, it's interviews with people to test whether they truly have the unmet needs you hope;

  • another day, it's spending a few hours "in someone else's shoes" to learn about what's happening in a part of their life; and

  • another day, it's asking individuals about one of your solution ideas to gauge their reaction.

As a result, what's best during exploration is a group that thrives in chaos, complexity, and environments of uncertainty, and meets it with curiosity and low ego.

When most people tell you to hire the "best", they mean the technical experts who win at being an expert.

And here's the rub: for many experts "if the only tool you have is a hammer, it is tempting to treat everything as if it were a nail" so their go-to will always be their technical tools-of-the-trade

However, what you need are people with the attitude that lets them do all the important things that are critical during exploration, especially when the important things aren't technical.

When your group is running, all of your group should focus on the highest-stake, critical questions to solve - regardless of their expertise.

When your group is recruiting, check how comfortable your specialists are with working outside their specialty, e.g.

  • will a software developer get on zoom with a nurse?

  • will a designer show out a just-good-enough design to a potential user?

If you get pushback, that may be a red flag for fit. They may fit better later when the context has changed to a more established, more predictable environment.

Lastly, this isn't limited to startups at all, it applies wherever you are trying to solve a problem you've never solved before - because the context requires adaptability, regardless of the org. chart or silos!

User Avatar

Steve Sanderson🚢

4y ago

Is Someone Rolling Out of Bed in the Morning Hoping For This?
Steve Sanderson @stevesanderson

If you're in any way involved in making and delivering products or services, you've probably felt the pull that new features are shiny objects e.g.

  • Customers ask us for a change - and tell us the feature to build

  • Technologists propose features they can build

  • Experts know how some feature will make things better

We've all felt the certainty, or the pressure, that this new feature will be right, but what if we're wrong?

One way we get there is by confusing means vs ends - we think our shiny new feature is the "end" for our customers when it's actually a means to some result they want.

As a colleague, as a mentor, or when doing my own work, I regularly run across this confusion, so I had to come up with a way to keep from falling in love with the shiny object.

One way I do this is an odd question to try and get them to move away from loving the feature and towards empathizing and understanding with who they serve, i.e.

"is someone rolling out of bed in the morning hoping for this"?

And the answer is almost always, after honest thought, "no"

A conversation might go like this

  • X: "what do you think about this new feature that generates attendance reports"

  • me: "is someone rolling out of bed in the morning hoping to see a report on attendance?"

  • X: uh, no...

  • me: "what are they hoping for, what's keeping them up at night?" X: ... they need to reduce the attendance issues

  • me: so is your "report on attendance" a means to that end, or is it the end they want?

  • X: uh, it's a means to that end...

  • me: OK, let's work backwards from the end they want: "reduce the attendance issues"

  • ...

Simple, and a bit of a surprise, it might help you, as it has helped me, to reorient on empathizing and understanding the people we serve, and enabling more opportunities for innovation.