slide

Getting past snake oil

Martijn Rutten
5 min read

If your startup creates truly disruptive technology, the first reaction of potential customers and investors is often:

Forget it. It’s snake oil!

Disbelief. It sounds too good to be true. In case of Vector Fabrics, we created tools to parallelize unmodified C and C++ code. Silicon Valley business angels referred to this as the “holy grail of embedded systems design”. So great! Transfer the millions, we thought. Alas, in the same sentence they said: “snake oil!”. Others have made the same claim and never succeeded. It’s too hard. It can’t be done. Snake oil.

The snake-oil verdict hurts any truly disruptive technology. There is always some friend or colleague at the customer that has worked in your technical field in the past and failed. “Why would this startup that nobody heard of be smarter than myself or my team?” is the typical, unvoiced, response.

In our case, any old-school programmer in the semiconductor space has worked on transputers and parallelizing compilers. So yeah, it can’t be done.

Many startups with great technology, bright minds, and a truly disruptive approach have failed due to this sentiment at their customers and investors. How to get past snake oil?

Don’t demo. Just don’t

Here’s what not to do. Don’t demo your product. It is not about how shiny your product looks or what great features it has. You need to prove that “it can be done” first. Show the end result to prove there is light at the end of the tunnel.

A product demo still does not prove to a customer it will work in his way more complex environment. In our case, a demo of how a video decoder can be parallelized still does not prove that the 8+M lines of code in the Chrome web browser can actually benefit from parallel execution. But if you show a lightning fast browser running on 4 processors, you have made your point.

So is the only way to get past snake oil to come up with widely recognized proof points and customer references? Well, yes. In a snake oil market, the burden of proof is on you.

When Coverity (static analysis tools) had to prove the benefits of static analysis, they analyzed the Linux kernel - and found relevant bugs. The open source community proved to be a great reference.

No cure no pay

Creating proof points can be a pretty time consuming endeavor. In our, case we did parallelize the Chrome web browser and gained a decent 4x improvement in battery life for key HTML5 benchmarks. But the cost of understanding the Chrome code and productizing our patches for a set of mobile platforms was very high.

Ideally, you avoid a high cost of developing proof points, free customer workshops, or endless trial periods. Without strong references or proof points, that may be hard to avoid. Any smart sales person will try to at least get some commitment up-front from the customer. You may try a contract with warranties, a money-back guarantee, or at worst an early-adopter discount. Such contracts require a strong relationship of trust. I you can leverage that, that may just tip the scales.

Foot soldiers and a Trojan horse

What can you do to make your story more credible?

If you can create an army of foot soldiers, a following of true believers in your products, one of these may turn the tide in that critical meeting with your big prospect. But how do you get foot soldiers? Get a messiah; hire a Steve Jobs. Or more down to earth, put something of real value in the hands of your target users. Something they use in their daily work. Something they can easily get access to, preferably free. Feed them new and valuable insights fromĀ  your blog. Ship a free product that adds real value, albeit with limited functionality. And of course that free product rocks. It establishes your company as one of the few software companies out there that make products that actually work.

However, your army may not be big enough to conquer snake-oil. You may have an impeccable reputation and a large fan base. If your customers simply do not believe his problem can be solved, you are still in the woods. No matter how much shiny technology you throw at them, the response is still: “it can’t be done”.

trojan horse Time to bring in the Trojan horse. The best way to convince a customer is always from within. Sub-contract yourself or preferably co-founder to the customer for half a year or more. Make sure she or he nails the job at the customer. Once inside, it is much easier to open doors and eradicate all fears of snake oil. Don’t make the mistake of sub-contracting your brightest employee though. What are the odds the customer makes an offer your employee can’t refuse? A sure way to lose your best employee. Sub-contract a founder if you can afford to.

Hunting for pain

Finally, there may be a more direct road to sales bliss, by taking Geoffrey Moore to the extreme. Find a niche where there is an acute pain that your disruptive technology solves. Yes, you will get the snake oil response. But if the pain is large enough, you may get the benefit of the doubt. If this business is bleeding, managers will try any medicine that claims to have the cure. That requires the hardest thing in the world of startups: finding customers with an urgent and strong pain that you can address.

And if you find such a niche, sell pain, make ’em cry for your help.

Author

About Martijn Rutten

Fractional CTO & technology entrepreneur with a long history in challenging software projects. Former CTO of scale-up Insify, changing the insurance space for SMEs. Former CTO of fintech scale-up Othera, deep in the world of securitized digital assets. Coached many tech startups and corporate innovation teams at HighTechXL. Co-founded Vector Fabrics on parallelization of embedded software. PhD in hardware/software co-design at Philips Research & NXP Semiconductors. More about me.