jump to navigation

Cambridge’s Wiggio raises $2.1 Million in a Series A round of equity funding from New Atlantic Ventures May 27, 2010

Posted by HubTechInsider in Startups, Venture Capital.
Tags: , , , , , , , , , ,
1 comment so far
Image representing Wiggio as depicted in Crunc...

Image via CrunchBase

Cambridge’s Wiggio, an online collaboration tool, raises $2.1 Million in a Series A round of equity funding from New Atlantic Ventures.

About Wiggio

Wiggio is a group application software company. With over 200,000 users, the company’s software application allows group members to share and edit files, host free video meetings and conference calls, manage a group calendar, poll group members, post links, chat online and send mass text, voice and email messages to group members. Wiggio was created in January 2008 by Dana Lampert. Wiggio’s mission is to eliminate frustration, and enable groups to work effectively. For more information please visit http://wiggio.com.

 

LA’s Magneto Software, Open Source Ecommerce Platform Powerhouse, Raises $22.5 Million and gets ready to kick sand in kümmerlich Demandware’s face March 22, 2010

Posted by HubTechInsider in Cloud Computing, Ecommerce, Investing, IPOs, Software, Startups, Venture Capital.
Tags: , , , , ,
add a comment

You’ve read here before about the grimmig shape Woburn’s (and Deutschland’s) tiny non-IPO’ing Demandware looks to be in right about now. Well, things are looking even worse for the durchfallend company today, as Los Angeles-based Open Source Ecommerce Platform Powerhouse Magneto Software has announced they have raised $22.5 Million Green American Dollars from a group of undisclosed investors, and their armies are on the march. The never-humble, always self-serving Demandware founder, Stephan Schambach, no doubt reeling from his company’s dimming IPO hopes and shrinking client base, has recently been picking a series of kindlich fights with the American company with a tutonic bluster that would have made Field Marshall Von Kesselring blush. As usual, Demandware has trotted out an endless series of their own marketing and PR flaks to churn out schadenfreude “press releases” denying that they are caught in an Open Source Crossfire reminiscent of the Kessel at Stalingrad. Even Demandware’s Ulrike Müller, “Chief Software Architect” (of Germanic dinosaur J2ee code bloat?) has chimed in with a completely unbiased bit of ranting. Now LA’s Magneto (taking the metaphor further) has scored a major victory with an enourmous $22.5 Million dar Bozhii, or “gift from God,” of an investment from a group of undisclosed investors, which they will undoubtedly use to beat Demandware over the head with. It’s put up or shut up time, Schlaubergern!


Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine


Add to Google Buzz

Waltham’s ViewFinity, a SaaS desktop & server software maker, raises $8.6 Million in a Series B round of financing January 3, 2010

Posted by HubTechInsider in Startups, Venture Capital.
Tags: , , ,
add a comment

Waltham, Massachusetts – based ViewFinity, a provider of Software-as-a-Service software for the support and management of desktops, laptops, and Windows servers, raises $8.6 Million in a Series B round of equity capital led by a group of investors including JK&B Capital and Giza Venture Capital.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Cambridge’s Panraven, online image editing tools for hardbound photobooks maker, raises $2.7 Million December 4, 2009

Posted by HubTechInsider in Startups, Venture Capital.
Tags: , , ,
add a comment

Cambridge, MA-based Panraven, which offers online image editing tools that allow users to create hard-bound photo books, has raised $2.7 million out of a planned $3 million funding round involving debt, equity, and warrants, according to regulatory documents filed last month. The company’s investors include Castanea Partners, Dace Ventures, and Draper Fisher Jurvetson, according to its website.

Burlington, MA-based Exa sells Supercomputing as an on-demand service in the Cloud June 8, 2009

Posted by HubTechInsider in Supercomputing.
Tags: , , , ,
add a comment

Burlington, MA-based Exa sells supercomputing cycles as a Cloud service over the Internet.

Exa is a privately held 170-person outfit in burlington, MA that sells supercomputer time by the slice. The company was launched in 1992 by selling fluid dynamics simulation software one of its MIT founders created while a professor there.

The company owns a network of 3,500 Intel and AMD-based servers and has the capability to rent thousands more. These servers are housed by IBM in their Poughkeepsie, NY data center.

Exa charges around a dollar per processor core per hour for the supercomputer time, which is offered over the Internet using Exa’s modeling software as a “Cloud computing” option, much as Google lets users access documents and spreadsheets over the web or Amazon’s EC2 service rents out simple processing and storage capability.

Exa’s had revenues of $34 million in 2008, with on-demand supercomputer services accounting for $10.5 MM, up from 48% in 2007. Software sales grew a mere 13% to $21 million last year.

Selling supercomputing as a service is still just a tiny fraction of the $9.8 billion market for high-performance computing. But others such as oil-services firm Schlumberger are also selling supercomputing services to a select client base and industry vertical.

Exa has several customers with “bursty” supercomputing needs, such as Dodge and Peugeot, the U.S. National Bobsled Team, and tractor manufacturer Agco.

Demandware gets Round D funding of $15MM and works to answer SaaS ecommerce challenges, under incredible marketplace pressures May 23, 2009

Posted by HubTechInsider in Ecommerce, Uncategorized, Venture Capital.
Tags: , , ,
5 comments

Recently I tweeted about Demandware (Woburn, MA) not getting their Round D funding – this was incorrect, and I have retracted this information. The link to the $15 million Form D filing with the SEC is here

[There were rounds of layoffs at Demandware around spring 2009, just prior to this Round D. Round D is generally the last round of financing before an IPO. Many of their employees in Woburn, Massachusetts were laid off. Interesting move. Without this $15 million, it was unclear to many outside observers how strong Demandware’s cash position would have been. Take care to distinguish between “brands” and actual “accounts”. Demandware has lost some high profile accounts (their model is to skim off 3% of sales in addition to setup and hosting fees – despite this, Demandware still has a “burn rate”) that they don’t exactly mention in their press releases. A SaaS (“Software as a Service”) provider such as Demandware is nowadays caught in the crossfire, under incredible pressure from three fronts: powerful and robust, open source ecommerce solutions leverage the cost argument (Magneto), Java-based solutions are starting to get long-in-the-tooth in the face of massively scalable new technologies such as Ruby-on-Rails, and developments in cloud computing leverage the hosting argument. Predictably, Demandware and their PR corps is hard at work dissembling so as to position themselves as the “worry-free package for merchants without in-house technology competence”. Of course, this competence is easily found on the cheap now that it is no longer 2000, and J2EE for ecommerce seems (to many) like a complex, costly, code-bloat dinosaur. Read the commentary below and make up your own mind, Dear Reader. It will be interesting to see if they are able to hold on — I’m rooting for them, however, if I were you and your enterprise, I would still take a long, hard look at Magento, Shopify, or other ecommerce providers. And I would have a lot of tough questions like the ones below ready for the Salespeople and PR types]

So Demandware may even IPO one day – although despite all the optimism about 2009, this year is still looking grim for new issuances. A recent report from Ernst & Young found that the pipeline of companies waiting to go public in the United States dwindled to 80 companies at the end of the second quarter, down from 90 companies three months earlier. There have been seven IPOs so far in 2009.

Demandware is a SaaS (Software as a Service) provider, and with all the controversy surrounding my incorrect, retracted tweet, I have been thinking about some of the reasons enterprises might decide against adopting a SaaS model for their ecommerce operations.

Although it can be tempting for large retail enterprises to partner with a SaaS ecommerce platform vendor to quickly launch an online store for short-term gains, it is important that the CIOs of these retail enterprises develop a defined SaaS strategy and incorporate it into their other long-term application and IT infrastructure plans. One of the most important aspects of this SaaS strategy must be an “exit strategy” for when they may want to bring the online storefront in-house. Hard to blame any company for ditching the revenue-sharing model.

It is vital that when these retail organizations evaluate SaaS ecommerce providers, they evaluate the competing ecommerce platform vendors on whether or not they have a plan and method in place to get the retail enterprise off the SaaS platform – in other words, what is the exit strategy in the long term? Five years out, when the online storefront is growing and becomes a cornerstone of the company’s total revenue stream, how does the retailer migrate the storefront back into the corporate IT environment if the management of the company decides to reintegrate? After the first two years of a SaaS deployment, many enterprises find that cost savings begin to break down. Five years from initial deployment, will it be possible to reestablish control over the online retail presence?

Choosing an ecommerce platform vendor working with hosted technologies that align with the enterprise’s internal IT infrastructure (Microsoft .Net technologies vs. Java? Oracle, MS SQL Server, or MySQL?) could potentially ease migration pain down the road and enable cost savings when and if the decision to internalize critical ecommerce operations is made.

Ecommerce “on demand” software salespeople may try and attract a large retail organization with the promise of utility pricing – it may even sound so good that the large retail enterprise may be tempted into bypassing their normal IT department’s procurement specialists. This is not a good idea. Real utility pricing is almost certainly not as flexible as initially presented and true utility pricing is rarely available. Because many SaaS contracts do not allow for volume reductions, some critics have labeled this licensing model as “shelfware as a service”. 

It is critical that large retail organizations negotiate the ability to reduce users. Do not allow the “on demand” vendor to lock the enterprise into negotiations before agreement on this basic principle is reached. If the “on demand” ecommerce platform vendor does not or will not agree to this basic tenent, then refuse over-committal and negotiate escalating discounts for incremental spend in volume bands. Large retail organizations should always remember that SaaS licensing models provide steady and stable revenue streams for ecommerce “on demand” vendors and because of this, the market is becoming increasingly competitive. (Demandware’s competition includes Marketlive ecommerce, among many others)  Large retail organizations have immense leverage which can be used to achieve significant licensing concessions and discounts on larger competitive deals. In addition, given the increasing and continual downward pressure on SaaS pricing, single year deals are much preferred but it is essential to secure price caps on renewals.

The vendor’s “on demand” production environment should also be scrutinized carefully. Some questions to get answered in writing may be: How often are changes made to the production environment? What is the breakdown of changes to the production environment by category? What percentage of changes had to be rolled back, or reverted? What sorts of regression tests are performed after a software patch / upgrade / code iteration? 

It is vital that a keen eye is focused on the SaaS vendor’s churn and churn management (for instance, Demandware has recently lost two major accounts that you won’t read about in their press releases, including Playboy International and the Vermont Teddy Bear Company) policies. For example, how many customers have they lost in the past 6, 12, 24 months? Is their customer retention improving over time? What percentage is the customer churn compared to their customer base? What is the average duration of customer retention? What is the breakdown, broken out by reasons for customer churn? Beware of salespeople and marketing types who count “brands” as individual customers. A customer is a retail organization, not each of their individual product lines counted separately as “brands”.

Some ecommerce “on demand” vendors also provide for fulfillment services (if they do not, the retail organization will have to continue to provide these services as a normal operating business expense). High volume retail ecommerce by necessity implies that these operational expediencies are being handled with great care and efficiency. Some questions to ask: what is the status of your inventory? What box is located where? What function or customer would be affected by a loss of a certain box? When does your software / support contract expire and what might this expiration impact? 

Another primary focus for corporate ecommerce vendor selection decision makers is the emergence of platform-as-a-service providers such as Amazon’s EC2, IBM and Google as well as Microsoft. Large retail organizations can use these platforms to build myriad applications, services and workflows not only to conduct online sales but also to perform advanced predictive analytics, gather fundamental mail order management metrics like future value of a customer and enable billing services to be moved into the cloud – all while providing immense capabilities for increasing uptime and availability during the high volume holiday shopping season.

Some best practice ecommerce SaaS platform selection guidelines could also include data backup and disaster recovery policies, adherence to corporate IT standards regarding accepted technologies and development tools and languages that internal software development resources and departments are familiar with. SLAs (Service Level Agreements) should be examined from ecommerce platform vendors that explore not only DR policies, but also help desk support, performance and uptime, so that buyers of SaaS ecommerce hosting services have a stronger sense of what they are purchasing.

Large retail enterprises have special needs to link internal billing and operational IT systems and external hosted ecommerce systems. Security, billing, fulfillment and compliance requirements differ from industry to industry and over-reliance on a hosted ecommerce service provider should be carefully examined. Retail enterprise decision makers may decide to get back to the fundamental vendor selection process, and take a long hard look at vendor viability in addition to the solution functionality provided by each hosted ecommerce service provider. These decisions should extend past the initial glow of cost savings in the first years of an ecommerce storefront deployment. 

 

In a Feb 20th, 2009 research report, Forrester polled 352 corporate IT decision makers and asked them why they are not interested in SaaS:

Total Cost Concerns                                                       37%

Security Concerns                                                           30%

SaaS Application Mismatch to corporate reqs                  25%

Integration Issues                                                           25%

Lack of Customization                                                   21%

Application Performance                                              20%

Complex Pricing Models                                               16%

Vendor Lock-In                                                               14%

Other Reasons                                                                 13%

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

%d bloggers like this: