Holiday Test and Learn

During a presentation to marketeers, Innohead did a real-time Test and Learn simulation. Using their smartphones, participants was asked to fill out a small survey. For instance, a question is about to what degree you "like to experience frightening things".

Respondents were then randomly exposed to an offer of going to either Burning Man-festival in Nevada, or, to go on a shopping trip to Milan, Italy.

It quickly turned out frightening experience-types were much more likely to want to go to Burning Man. Testing can be easy.

Financial Institution Adopts Narrative Sign-up Form Based on Innohead Advice

Nykredit, a Danish financial institution, was advised by Innohead to try out another type of sign-up form. Experiments have indicated that so-called narrative sign-up forms may yield higher sign-up rates. One study at, quotes increased sign-up rates in the range 25-50%.

Nykredit's narrative sign-up form is shown below.
The title is "Become a Customer in Nykredit Bank". The text reads "Hi, my name is [name] and I would like to hear more about switching to Nykredit Bank. I live at [address] in [zip code] [town] and can be reached on [phone no] or via [email]".

A smart website owner would create the new style form and then A/B test it against the old style.

T.Rex in the Canteen

One of our clients has an awesome canteen. When eating there, tangible ideas about the food chain and your place in it somehow drips into the mind.

Innohead Is Growing

The new logo
Innohead is growing: We welcome Rune Haahr Petersen onboard. He holds a MA in History, wrote about Danish Missionaries in India, is smart and sings bass in a choir. His idea of fun is to visit The National Archive and record the historical prices of grain auctions to study the price transmission mechanisms. Rune is initially focused on servicing a big client in the transportation sector, that needs to get their marketing intelligence infrastructure up to speed.

The revamped website launched on using the new logo from Eagle Logos and a crowd picture from Istockphoto. Over time we might swap in pictures of street crowds, city lights seen from the sky, satellite pictures of houses and so on. Oh god, how Internet Explorer can be a pain when designing websites.

We are experimenting with a Reinforcement Learning-based marketing campaign manager. A bit like the Tic Tac Toe-player, it learns from its mistakes and successes. Imagine focusing on email frequency: If you email someone too often, they will become irritated and unsubscribe. On the other hand, if you email too rarely, they will start to forget you. Experimenting with various email frequencies on an automated basis, the Reinforcement Learner will conduct a large number of experiments and learn from these continuously. We are in talks with a NGO to allow us to play with their donor base using this approach.

Nothing is certain yet, but we hope to land a mobile network operator as a new customer.

A Test of the Recommender Engine

Score! The first results from The Product Recommender Engine arrived.

A group of customers received a business-as-usual email of product highlights. Some of these customers returned and bought some more stuff. Let's call that conversion rate index 100.

Another group received customized emails, recommending specific products, based on the product purchase history as decided by The Engine. These customers converted at a rate of index 157. Money was made.

Innohead is now in the process of packaging the recommender algorithm as a standard component on the platform. I wonder who else, besides booksellers, could use the system?