I have sat in a lot of pitches over my career. In advertising, we are in the business of selling ideas. It is no different in VC-backed deep tech. I know what makes an excellent pitch, and I know what makes a painfully cringeworthy one.

It is not purely down to the tech. It is not purely down to presentation skills either. So here are six things you can avoid to put yourself ahead of the pack.

1. Leading with technology, not the problem

You built something remarkable. You want to show it off. So you open with architecture, with specifications, with clever engineering that took eighteen months to solve. Worse, you assume the investor already understands why it matters, because you have lived inside the problem for years and they have seen forty pitches this month.

Compare: "We've built a photonic sensor array with sub-nanosecond latency" against "Right now, a soldier waits four seconds to know if a signal is a threat. We've cut that to under one."

Nobody cares yet, and nobody follows automatically. Show them the wound before the cure, and spell out the advantage in language a smart generalist can follow in one pass.

2. Confusing what you built with why it matters

"Our sensor fusion reduces false positive rates by 60%" is a fact. "We cut a threat-detection process from six hours to six minutes for an AUKUS-aligned customer" is a reason to invest.

Same pattern, different domain: "Our quantum sensor detects magnetic fields at the picotesla level" is a fact. "We let a submarine find what a satellite can't" is a reason to invest.

Facts describe whereas outcomes persuade. Take the time to know the difference, and always lead with the latter.

3. Spending your detail budget in the wrong currency

You are passionate, I get it. But here is the thing most founders get told wrong. You are not being too detailed. You are being detailed about the wrong things.

You have a finite amount of attention in that room. You spend it on the part you find most interesting. The architecture. The physics. The eighteen months of engineering. Meanwhile the investor is waiting for a number they can actually use. What it costs. What it replaces. What the customer pays today and what they would pay you.

So you walk out feeling thorough. They walk out feeling starved. You are both right.

I have sat through so many presentations where the first 60% is all technology. Half a second looking around the room tells you everything. Everyone has switched off. Not because there was too much, but because none of it was in a currency they could spend.

Remember investors buy the jockey and the horse. But at first contact there is no jockey in the room. There is a cold email and a deck, and every slide has to earn its place on its own.

So cut ruthlessly, but cut in one direction. Strip the physics back to what a smart generalist can follow, then spend everything you save on commercial detail. Push the rest to an appendix.

If a detail does not change the investment decision, it does not belong in the room. If it does change the investment decision, it had better not be buried in the appendix.

4. Treating the market size slide as decoration

There is a difference between stating a market size and showing an opportunity. A total addressable market number is a data point. An opportunity is a plan to capture something specific.

Too many founders paste in a generic TAM figure, sourced from a report nobody reads, and move on. "The global defence market is worth $2 trillion" tells an investor nothing. "We win the $40 million sensor-fusion budget inside AUKUS Pillar II first, then expand into allied navies" tells them everything.

Optimise for capitalising on the opportunity, not for stating the number. Investors know a lazy market slide when they see one. Show your reasoning. Show the specific wedge you will win first, and how it expands.

5. Failing to show traction as proof, not just progress

There is a difference between proof of traction and a status report of your sales pipeline. One is evidence. The other is admin.

"We have signed three pilot customers" is effectively a status report. "We have signed three pilot customers, and two have already expanded their contracts within ninety days" is proof.

The first tells an investor you can sell. The second tells them the product sells itself.

By all means outline the strategic conversations and partnerships you are having. But say why each one matters. Who is having conversations on your behalf when you are not in the room? Which relationship unlocks a door you could not open alone? That is the part investors are actually listening for.

6. Ending without a clear ask

Arguably the worst of all. The pitch builds well. The story lands. Then it fizzles at the end with a vague gesture toward "future plans" instead of a specific ask.

State exactly what you want. The amount. The use of funds. The milestone it unlocks. Ambiguity at the close undoes clarity built earlier.

The pattern behind all six

Each sin traces back to the same root cause: founders describe what they built instead of translating why it matters. Technical strength does not sell itself. It has to be interpreted, simplified and connected to outcomes the listener already values.

This is not just my read of it. There are now published teardowns where working VCs mark up cold pitches line by line, and they land in the same place. The pitches that fail do not fail because the idea is weak. They fail because the investor cannot construct a commercial case out of what is in front of them.

I call this the commercial translation gap. It is the distance between what a company can do and what the market understands it can do. Deep-tech founders are usually brilliant at the first part. The gap is where the value leaks out, and it is almost always wider than the founder thinks.

Close the gap and nothing about the technology changes. Everything about how it is received does.