As a software company you might think that we tend to write all our own internal business applications to manage finances, customers and projects. However, just like finding time to update the website, we find ourselves immersed in our core products and business software solutions. As a result we go shopping from time to time...

One such occasion happened this week after we sought to improve the management of sales leads. Having used TactileCRM for a while we really found that it wasn't quite doing the job, so after a few searches into Google for "web based simple CRM software" and trawling the results I put some thought into what criteria made a good business decision. The resulting a SaaS business application scorecard of sorts but before looking at this I'll recap on some background SaaS products.

Why are we looking at using a SaaS product?

First and foremost we want a web application product. With our background in rich internet application technology we release the power of information by joining everyone together on the same platform. Improved information sharing and communication follow naturally.

As any reasonable (and cost effective) web based product will be a software-as-a-service model. Furthermore a fully outsourced model with hosting, backups and upgrades all taken care of makes using SaaS a no-brainer.

So what are the risks of using a SaaS product?

Putting your organisation's mission critical data onto remote web application which you do not control is inherently a risky business. Without access to the database, backup facility and often little in the way of a personal service this is amplified. Furthermore no control to customise the functionality - even at a cost - means that, as though the Internet blogs might claim, SaaS is no enterprise software killer (interesting article here, http://msdn.microsoft.com/en-us/library/aa905332.aspx).

The risks are potentially huge and so the main classification on the suitability of the product depends on the value of the information being stored. For our lead management problem this ranks up there with financials and source code.

A summary of scorecard factors

After some not too much deliberation the following factors were included in our scorecard for the purposes of finding a decent web application to manage leads...

As a result we came to the conclusion that there were three overarching and imutable factors that played a key part in our scoring.

The API (Application Programming Interface)

Without the ability to connect to the SaaS solution and draw out data on our own terms we simply were not going to consider the solution. The API means that even if features are missing we could write them in our own code. However, this depends on...

A good data model

No matter how advanced the API a bad data model will give problems down the line. So what is a "bad data model"? Ah, well this is perhaps where you need some consultancy but our criteria for a good model are:

• Abstraction between application components
• Separation of concerns
• Arbitrary and extensible object properties
• Searchable abstract object properties
• Sharding of functional components
• Clear naming conventions

The most common issue we found was the fairly innocuous "Searchable abstract object properties". This appears to be an inherent problem with software design; putting data in is the logical first step and retrieval always comes second. The net result is that solutions tend to focus too much on the storing of data and not enough on the retrieval when this is the very reason we are storing it in the first place.

Service Level Agreement

Miraculously not all the SaaS providers make their service level agreements very readable or clear. SalesForce and Zoho, being larger organisations, have terms of use that cover all their services and so shed little light on what happens if you would like to get access to your data.

Looking at Tactile and Capsule there are some interesting, and perhaps frightening, points in their Terms of Use:

From TactileCRM (http://www.tactilecrm.com/terms)

We shall not be liable for:

11.2.2 any loss or corruption of Your data arising directly or indirectly as a result of You using the Service;

From CapsuleCRM (http://capsulecrm.com/terms/)

...exclude all liability and responsibility for ...including without limitation, any direct, indirect, punitive or consequential loss or damages, or any loss of income, profits, goodwill, data, contracts, use of money, or loss or damages arising from or connected in any way to business interruption, and whether in tort (including without limitation negligence)...

Interestingly those are both English companies so it would be interesting if the indemnification of liability would actually stand-up in a law suit.

All providers fail to offer an uptime guarantee or user-controller backups. This is perhaps one of the core vectors of differentiation that they should all capitalise on and something that Byng Systems feels is essential to any hosted solution.

So, to the winning sales lead tool...

Finding the right solution was no walk in the park as there appeared to be issues with all the products trialled. This shortlist is...

• 1) SalesForce (the original SaaS CRM, way too much functionality)
• 2) CapsuleCRM (web based CRM, have some SEO guys :)
• 3) Zoho CRM (a good platform, but like SalesForce, way too much)
• 4) Tactile CRM (our existing platform, lacking data model)

We found that CapsuleCRM topped our scorecard and met our needs the best. This is based on our requirements which resulted in SalesForce and Zoho being way too overcomplicated.

Check them out for a decent solution a simple online CRM. One feature that would make is amazing would be a Lead Worklist... this would make a good candidate for something for Byng Systems to develop in our labs