Software as a Service has been with us for a while now and the initial hype has died down (only to be replaced by Cloud Computing hype, although this is moving out of its own hype phase and has to start delivering on its promise now). But what is the real uptake of SaaS and is it the future?
SaaS has been hugely successful in some areas, and with good reason. The most obvious examples are email (Google, Microsoft) and CRM (SalesForce, Oracle) where the promise of low monthly costs per user and the removal of all the systems management work have appealed to a lot of people. Where a very standard, functional service is needed that doesn’t deviate too far from the norm, organisations (particularly SMEs) will turn to SaaS to avoid upfront capital costs and get a solution in place very quickly. Sometimes these decisions are also made to bypass internal IT departments who are perceived to slow down the procurement process.
However, SaaS, at least in its current state is not the answer to everything. Most offerings provide relatively little opportunity for significant customisation, and integration with other systems is often very difficult and/or limited. If your IT systems are meant to be a differentiator and give you competitive advantage, then SaaS doesn’t appear to be a viable option in many cases. Other concerns also linger about the privacy/security of data and the danger of SaaS providers disappearing – it isn’t much of a stretch in the current economic environment to envisage a SaaS provider suddenly stopping its service. If that happens then the backlash on all SaaS providers could be significant.
SaaS is clearly going to be with us in the future, but I think it’s a long way from being the future of all IT solutions. For more commoditised types of applications then it’s a great idea but I don’t think many companies (beyond the very small and straightforward) are going to be going 100% SaaS just yet.
Andrew
I've spent the last few days pouring over statistics for one of our clients websites, as part of a periodical review of the site. The aim is to try and identify trends, changes and improvements to the site since it originally launched. In that time, there have naturally been many changes and tweaks, and it's useful to be aware of how you can monitor the effect of these changes.
We recommend Google Analytics for all our clients who have public facing websites. Analytics is a great bit of free kit from Google that tracks a massive amount of data about your site visitors, where they come from, and how they use the site. If you run any online advertising campaigns with Google, Analytics will track additional data about the results of that as well. Analysing all this data can be quite time consuming, and with the vast array of information that gets collected daily, it's easy to make incorrect assumptions about trends and changes without taking into account a lot of factors.
But if you're willing and able to invest the time, the rewards can be great. In this particular example, we've been able to identify a raft of possible changes we think the site would benefit from, and together with the client, will be rolling this out over time. The thing to remember is that, with these changes, it's important to monitor their effect. Sometimes a change will produce totally unexpected results. Sometimes it'll make no difference at all. The important factor to consider is "how will I measure this". There's little point making a change if you can't directly measure it's impact.
In the case of websites analytics, that is perhaps the most useful and telling data you can collect. A website is a living entity, that should evolve continually. You can draw any number of conclusions from reviewing historic data, and as the title suggests, you can come to whatever conclusions you like, when choosing the correct data. Only by making changes, and monitoring the end results, will you truly build up an accurate picture of how your site is performing, and what drives your visitors to keep coming back.
Something to think about next time you're reviewing your weekly traffic stats?
Paul Dunlop
e: 