At yieldHUB, we pride ourselves on listening to our customers, users and prospective customers. Looking for a title for this blog we checked the definition of the phrase “to lend an ear”. According to the Collins Dictionary, the phrase means to listen to someone carefully and sympathetically. The definition goes on to mention that advice is then offered. That really does fit with how we set out to work at yieldHUB. We listen, carefully and sympathetically, we then offer advice and add to our roadmap as appropriate. 

This article is in part written to highlight the way we work at yieldHUB for prospective customers and also to remind our existing customers of the value in sharing feedback, ideas and suggestions. 

We’re always looking for the high-value items which add to yieldHUB without overcomplicating or detracting from the core functions. It is an interesting and highly responsible position to be able to take ideas and input from many people, look for common elements in problems and find solutions that can solve many problems for many customers. 

High-value functionality

A great example of this is the recent addition of multi-test charts. Historically we didn’t include multi-test charts in yieldHUB as there are inherent risks in combining different data into single plots where the source data could be in completely different units or scales. With growing interest from customers in this feature, we added this to the user community page which described other customer’s requests and the potential pitfalls. This experienced engineer was then able to define a carefully thought out specification for the charts that was an excellent compromise to provide the majority of the functionality with a minimum of risk. 

Following implementation this senior engineering manager replied to our automatic survey (sent every time a ticket is closed) with the following statement “Multi-test histograms are so useful, the implementation is really great and I find that I use them all the time.” So, we listen, we develop, we implement and then close the loop by asking for feedback on every ticket closed. 

“As we grow our team of product and test engineers, with yieldHUB’s ongoing commitment to supporting and improving their software, we are confident that it will continue to meet our ever-evolving needs.”

Ray Clancy, Vice President of Product and Test Engineering, Movandi

Increased Efficiency

One of the things that we’re always looking for is an idea that helps to make our user’s working lives more efficient. One such example has been to automate the setup of characterization. During 2019 we developed an extensive characterization module. As our customers ramped up the use of characterization we listened to the feedback. The early stages were to add more options to use data generated in different ways with conditions defined in different ways. The potential problem then is that there are many options as to how to prepare the data before generating the report. So we added an option to automate the preparation of the data. Set it up once (which can be a guided setup with one member of our customer success team) and then re-use that setup in two clicks of a mouse.  

In this case, we reduced the questions from customers using the characterization module for the second or third time from about 1/5 to almost zero. To say that another way, one user needed to be reminded how to setup characterization for their data. Now yieldHUB remembers it so the user doesn’t need to. Therefore the questions from users about this have almost completely stopped. In short when a user says to us “Wouldn’t it be great if yieldHUB could remember this so that I don’t have to?” then our answer is “Yes it would be.” and we deliver on that. 

“You can tell that it was developed by people in the industry, who know what they are doing.”

Patrick McNamee, Director of Operations, EnSilica

Sometimes it’s the Small Things

While everyone loves a new idea that adds something different, shiny and new to yieldHUB we also pay attention to the detail of some of the seemingly small requests. 

For many years our customers used our genealogy correlation tool to find relationships between the data from different manufacturing stages. This is presented in a summary table with R factor correlation values. Following a customer suggestion, we changed this to R squared so that the strongest correlations can be quickly ordered to the top of the table regardless of the direction of the correlation. 

With regards to yieldHUB tools, we provide analysis “From high level to die level”. That Die level analysis tool has also been enhanced. The options now include showing the site number that a die was tested on and the option to select which “incidence” of testing to show. This is particularly useful when die are re-tested. By default, the last result per die is available in analysis but it is now possible to select the first result and analyse that in detail using the Dice Analysis tool. 

“yieldHUB’s big data software and worldwide support has allowed us to manage our yields effectively over the years and their solutions have evolved with our needs.”

Julie Holland, Vice President, Corporate Operations of Diodes

Summary

In summary, managing yieldHUB’s roadmap is an interesting and varied task. By listening to our customers we strive to continuously add value. We love the feedback that we get, here is another quote from a user responding to our customer service survey “The team are very responsive, helpful and nice to deal with”. 

Finally, we would like to thank the users who take the time and effort to share ideas and provide feedback about yieldHUB, we are building this system together with you.