Pivoting

Sometimes the evidence clearly says no — and then you don't need to give up, but to change course. That is called pivoting. A pivot is not a defeat; it is one of the most common routes to a model that actually works. This final lesson is about changing direction without throwing away everything you have learned.

What a pivot is — and isn't

A pivot is a deliberate, fundamental change in one or more building blocks, while you keep what you have learned. Note the words: fundamental and keep the learning.

A pivot is not changing the color of your logo or adjusting a price — that is ordinary optimization. And a pivot is not giving up and starting something completely unrelated from scratch either. The real pivot keeps one foot in what you already know: you take the insight, customers or technology you have built, and twist the model in a new direction the evidence points to.

Common pivot types

Pivots come in many forms. Some of the most common:

• Customer-segment pivot: the product is good, but for the wrong group. You keep the solution but aim it at a different segment.
• Problem pivot: customers like you, but care more about a different problem than the one you solve. You follow the pain.
• Platform pivot: you go from being an app to becoming a platform others build on — or the other way around.
• Revenue-model pivot: the product lands, but the way you charge doesn't. You switch, for example, from one-off sales to subscription.
• Channel pivot: you reach customers in a completely new way, which changes the whole model.

Kolonial is again an instructive example: they started out wanting to sell technology to others, but pivoted to running their own online grocery business and became Oda. They kept the technology expertise but changed customer, value proposition and revenue model.

The point is not that everyone has to turn that dramatically, but that a pivot usually changes several building blocks at once — as long as the change is grounded in what the evidence actually tells you. Small, random course changes with no basis are not pivots; they are just drift. A real pivot is a considered turn based on something you have learned, not a new idea you grabbed because the last one got boring.

Signals that you should pivot

How do you know it's time? Look for patterns, not single events:

• Growth that stalls no matter what you try.
• Customers using the product in a completely different way than you intended — often for one specific thing.
• A small part of the offering that people love, while the rest leaves them indifferent.
• Unit economics that will never add up, however much you fine-tune.
• Repeated experiments giving the same negative answer.

The last point matters: one failed test is not a signal to pivot — it is a reason to adjust. But when several honest tests point the same way, it is time to take the signal seriously.

Pivot without losing the learning

The danger of a pivot is to "reset" and discard everything. The wise pivot is economical with what it keeps. Before you turn, write down three things: what worked (keep it), what didn't work (discard it), and which insight about the customer still holds (build on it).

Carry the customer relationships, the technology and the hard-won insight into the new model. A pivot is not starting over from zero — it is using everything you have learned to make a wiser choice than you could have made yesterday. That way each lap around the build–measure–learn loop is a step closer to a model that holds.

Do this now

Look at the results from your experiments in the previous lesson. Judge honestly: does the evidence point toward continue, adjust or pivot? If a pivot is tempting, draw a table with three columns — "keep", "discard", "build on" — and fill it in. Then you have a pivot that carries the learning forward, not one that starts from zero.