Business Leader Interview - John Symon talks to Peter Wren-Hilton, CEO of Pingar

In a recent visit to Australia & NZ, John Symon, SVP Asia Pacific for Document Boss, met with a number of innovative ECM related software companies that are now beginning to expand their penetration into overseas markets, notably in North America, the UK and specific European countries. Whilst there, John met with the directors of the NZ based software company, Pingar, and gained some insights from CEO, Peter Wren-Hilton, on their solutions and growth plans for their business.




1. Can you share with us some background on yourself and what motivated you to start Pingar?

I arrived in New Zealand from the United Kingdom in 2003. Having spent 8 years working for Butterworth Law Publishers as a Managing Editor, I established HB Internet in London in the early 1990s to build database driven websites that would enable enterprises to take advantage of the emerging Internet. In 2001, I established HBI Software in Bangalore, India. This provided some external development support to HB Internet, but its main purpose was to begin to develop its own, web-based IP. It gave me my first direct experience of building and managing an offshore business.


The main motivation for establishing Pingar in 2007 was an increasing frustration with my inability to find relevant documents quickly, using conventional search technologies. I thought there had to be an alternative to plain keyword search. It took four years of research with key University collaborators, in both New Zealand and the United Kingdom, to build the algorithms that today drive Pingar's document understanding technology.


2. What products and solutions does Pingar offer and what are the main end user challenges they resolve?

Pingar offers an API which contains a number of different components to help enterprises understand their unstructured data. Pingar's most popular product area revolves around our algorithms’ ability to read and understand document context like a human. In the Microsoft SharePoint space, this has enabled Pingar to produce two Entity Extraction tools for SharePoint 2010 & 2013 which automatically scan documents for keywords, taxonomy terms and known entities such as people's names, organisations, locations, addresses, dates etc. and then automatically extract and assign these entities as metadata to the documents as they are uploaded to SharePoint. This provides consistent metadata across all documents and is particularly valuable when large numbers of documents are migrated into SharePoint from file shares or other document management systems.


The main end user challenge this resolves is the abolition of the need for users to tag documents as they are uploaded to SharePoint. Most users don't or will tag only to meet the minimum set requirements. As a result, many aspects of SharePoint's functionality fail to deliver, leading to a lack of user adoption and a poor return on capital investment. With comprehensive metadata, you can create search refiners, build e-discovery applications and add many value-added workflows & processes to your SharePoint implementation. Without comprehensive metadata, you sometimes only end up having an expensive file share. Pingar's Metadata Extraction tools for SharePoint are designed to avoid that.


3. How do you position yourselves in the market versus any competitive products? What is Pingar's "magic sauce" ?

Pingar's API is a web service and as such is platform agnostic. Whilst our immediate target Go to Market has been Microsoft SharePoint, our metadata generation technology can be applied to any content or document management system. This is playing out today in the broader 'Big Data' space where a number of global consultancies are building user cases using our API to address major, unstructured document challenges in very different domains: medical & pharmaceutical, legal, banking/finance/insurance, media & publishing are just a few such examples.


The secret sauce really is the deliverables from the four years’ original research which continues to be an on-going process and investment. Underlying our API components, which were released last year, is a combination of natural language processing & computational linguistic technologies. Each component has then been packaged into a machine learning framework. It is this patent pending technology that enables us to continue to develop real products and solutions that tackle the on-going enterprise challenge of managing growing volumes of unstructured data.


4. As a NZ based start-up technology company, what have been the biggest challenges in gaining awareness and traction in the broader international market, and how have you addressed these challenges?

From Day 1, we knew that, in order to scale our business, we had to play on the global stage. Today, we have offices in the United States, the United Kingdom, India, Hong Kong & of course, New Zealand. We are fortunate to have recruited a very experienced, international management team to manage these offices and the marketing & business development teams in each of these regions.


One other aspect, of course, was to recognise that the challenge of unstructured data is not just an English-language one. Over the past eighteen months, we have invested considerable resource into making our algorithms multi-lingual in terms of document understanding. This has definitely helped increase our profile across Europe as our algorithms can now read and understand documents in German, French, Spanish & Italian, while our ability to understand both traditional and simple Chinese language documentation is a major plus for our Hong Kong-based team. Over the coming year, we will be expanding our multi-lingual capability significantly.


5. Clearly your solutions impact multiple areas of the ECM and related technology & applications markets with the ability to automate the capture, classification and search of metadata, giving meaning to unstructured information. Which target markets and applications are you focusing on first and why?

Our first target market is the Microsoft SharePoint eco-system. Our second is the scanning space.


As the question implies however, there are just too many target markets and applications that our core API components can address; the challenge of unstructured data is not diminishing in any way. We therefore, recently launched a Better Together Partner Channel program to identify best of breed partners across a range of domains. I think our strategy to observers is pretty clear. We intend to focus on building new functionality on top of the API components we already have to allow other software developers, system integrators & consultancies to build custom applications for their specialist domains. At the end of the day, Pingar is not and does not intend to become a domain expert. Our strength is in our technology. Making that available to third party software developers, system integrators & consultancies is therefore a key goal for us.


In terms of application and product development, we recently completed a 12 month research project with the University of Waikato in New Zealand that sought to build a taxonomy engine that could generate automatic taxonomies 'on the fly' by analysing individual document sets. Using that research, we have built a prototype engine that today is building such taxonomies. We expect to release a beta version of a packaged engine to the market in Q1, 2013. This brings natural language processing and machine learning to what has traditionally been a very manually intensive function. We are very excited by the opportunities that this breakthrough technology will bring and are keen to get more specific market feedback.


6. I understand you have just participated in the Microsoft SharePoint 2012 conference. How does Pingar's technology integrate with and add value for SP users?

For SP users, the fact that metadata no longer has to be manually tagged to documents as they are uploaded to SharePoint will be widely welcomed. I think most SP Administrators will be leading the applause! When we first demonstrated this technology at the Australia, South-East Asia & New Zealand SharePoint conferences earlier this year, Pingar was voted 'SharePoint Idol' by delegates.


At the SharePoint Conference in Las Vegas earlier this month, we were able to announce Partnership Agreements with BA insight, Nintex & Stealth Software.


BA Insight will now be able to improve the search experience for business users by automating the analysis of documents to create meaningful context for classifying, navigating and quickly identifying their significance. BA Insight's Longitude products and TotalView solutions will be incorporating entity extraction and content analysis to accurately summarize documents, assign metadata as well as redact and sanitize documents on the fly.


For Nintex, by optimizing enterprise search, enabling businesses to analyze, organize and summarize the information contained within large numbers of documents, Pingar's API technology will extend Nintex SharePoint workflow processes, providing customers with further business efficiency and productivity.


By bringing entity extraction to the Stealth Content Store, Stealth Software is extending additional functionality to its users. In a better together scenario, organizations will benefit from greater user adoption, improved search and the ability to incorporate secured content into structured business processes. Stealth's ability to collect and secure large amounts of content, along with Pingar's ability to understand unstructured text, also introduces a new opportunity for the Big Data market.


These are three great examples of how significant, third party software developers & system integrators are utilizing Pingar's API technology in the SharePoint marketplace and beyond.


7. Are you able to integrate with all leading ECM &EDRMS software platforms and repositories?

Absolutely. We are currently working with third party system integrators who are planning to do just this. Our Better Together Partner Channel program is designed to support partners building such connectors and to ensure that our technology can benefit all leading ECM & EDRMS software platforms and repositories.


8. Is your current sales model primarily via a reseller channel, direct or a combination of both?

Initially, we have generated some direct sales for validation and reference purposes. This is not however, our long-term GTM model.


Our real focus is on building the Better Together Partner Channel program. If any readers are interested in learning more about how they can join this program, the quickest and easiest way to do this is to email me at The more information you can provide me with about relevant domain or technology expertise, the better. Our Better Together Partner Channel program is designed to provide real revenue opportunities for partners as well as providing them with the opportunity to build their own domain specific IP.


9. Which geographies do you currently cover and how are you able to provide 24x7support?

As already indicated, Pingar is formally represented in North America, UK/EMEA, Asia and Australasia. We provide normal office hours support in each region.


Currently, we provide 24/7 email support via our offices in Bangalore. We intend to launch a 24/7 person to person technical support service for our Better Together channel partners in Q1, 2013. Given our geographic representation, we operate very much as a 24/7 company. Extending this formally to our Better Together channel partners is an important part of the evolution of this program.


10. How do you see the rise of SaaS, Cloud computing, Social Media Communications, mobile and Big Data impacting the growth of Pingar?

It's fair to say that each of these represents significant growth opportunities for Pingar.


If we go back to basics, we know that 80% of all enterprise data is unstructured. The amount of enterprise unstructured data is growing exponentially at 40% a year. The growth of SaaS, Cloud Computing, Social Media, mobile and Big Data will only exacerbate the challenge enterprises have in understanding just what value is locked inside this data.


Pingar not only analyses unstructured data in order to generate metadata. We can also analyse unstructured data in order to extract entities and map them against fields in a database such as SQL Server or Oracle. You can then begin to build structure out of your unstructured content and analyse it via appropriate front-end applications & dashboards. The incremental growth in unstructured data volumes is forcing CIOs to find new ways of unlocking it. SaaS, Cloud computing Social Media, mobile and Big Data are simply driving that growth.


11. What other market drivers are you seeing now and expecting in the future?

Given the particular economic times we are in, enterprises are looking at new ways to get better value from the investments they have made in their ECM & EDRMS software platforms and repositories. The emergence from University research departments of natural language processing, computational linguistics and machine learning technologies that are now available in the public arena as real enterprise platforms and products, offer significant opportunities for improving productivity and reducing costs.


At the recent Text Analytics World Conference in Boston, Dr Alyona Medelyan, Pingar's Chief Research Officer, gave an insight into this future. Running Pingar API instances across an Amazon web service configuration, we analysed and applied metadata to 1.7M documents over one weekend. Had a single human being undertaken the same task, it would have taken 43 years.


Currently, the vast majority of enterprises rely on humans to tag their documents; the opportunities for automating such manual tasks by leveraging these emerging technologies are vast. As I have said already, this not only speeds up the process, but the existence of consistent metadata values across data repositories generates improved user adoption, creates better search capability and allows for improved workflow opportunities across your ECM and EDRMS platform.


12. What are the three biggest challenges you face over the next 2-3 years in order to achieve success?

The biggest single challenge remains that of education. Most enterprises, even ECM & EDRMS professionals, are simply not aware of how far technology has come in these areas. When we demonstrated the Pingar Metadata Extractor for SharePoint 2010 & 2013 in Las Vegas, the most common reaction from delegates at our booth was 'WOW!'. System integrators and IT administrators simply did not know that it was now possible to automate many traditionally manual tasks through the deployment of NLP and machine learning driven algorithms.


13. What do you consider are the essential ingredients in building a world class IT software business?

There are several, but assuming that an innovative technology, a clear customer need and sufficient funding are in place, my main essential ingredient would be the team. As CEO of Pingar, I regard recruitment as one of my most critical functions. I am more than happy to admit that my aim is always to recruit better qualified people for key positions than myself. That allows me to delegate and focus increasingly on strategy.


The other essential ingredient is the global focus. This is especially important when you are based in a country like New Zealand. It is a great place to live and generates more than its fair share of innovative and exciting technology businesses. The main challenge that a New Zealand base poses however, is the size of its domestic market and its distance to market. In order to build a world class IT software business from any small country, a global focus is required from Day 1.

For more details on Pingar see