CRM six months on...

There’s an awful lot of hype in the world of CRM, and it can be a pretty insightful, and potentially humbling, exercise to re-visit a project to see whether you’ve genuinely added value to a client’s business. I spent a morning last week reviewing how a client had fared since their system was rolled out in January, and since our last formal progress review in May. The feedback proved to be pretty startling.

The project in question was focussed on improving the operation of the customer support function, and our involvement covered planning the project, defining requirements, and project managing the implementation.

When we initially helped them shape the business case for the project, it was pretty clear that by providing better visibility of their performance against service levels, they would be much better placed to tune their quality of service. Eight months after live, this proved to be accurate though, it must be admitted, they were significantly outperforming our expectations. The SLA performance statistics generated by the new system were now religiously studied by the management team, and had become one of the key areas of focus of the weekly review meeting.

When the data was first reviewed, it was apparent that some aspects of performance were falling outside their published standards. The company was now able to quickly identify and target problem areas, and by the time of our meeting the number of overrun calls was well under 1% of the total. Because of the commercial arrangements the company had place, this improvement would repay the initial investment in the system (including our involvement) over twelve times in year one….which didn’t seem to me at least too shabby a return on investment. The client also flagged a host of softer benefits, which I’ll cover in a later post, but from an SLA perspective the project, confirmed the wisdom of W. Edwards Deming’s observation - 'What can't be measured can't be managed, and what can't be managed can't be improved.'

The foundation for successful CRM...

Michael Bosworth/John Holland make the following observation in their book ‘Customer Centric Selling’ – ‘Because of the predominant perception that selling is more art than science, few companies have sales processes that traditional sellers can execute. We believe that this deficiency is the single most significant factor contributing to the disappointing results achieved with sales force automation (SFA) and customer relationship management (CRM) systems.’

Leaving aside the art v. science debate, this observation hits the nail on the head. High pay-back CRM systems are those that effectively automate good business process, whether it’s sales process, lead generation, lead tracking, or any of the myriad of key business processes that CRM technology can impact.

This isn’t how organisations generally implement CRM technology. Lured by the vendor siren call of quick results for a fiver a month, the process part somehow gets forgotten. It’s not that the resulting system necessarily fail to provide any benefit, so much as they only provide a small fraction of the achievable benefit.

Take the sales process highlighted above - CRM technology gives the sales manager new capabilities. Gone now are the hours of compiling multiple slightly differently formatted Excel spreadsheets. The forecast can be viewed in real-time, with graphical charting, and drill down capabilities, so that the sales manager can immediately validate the underlying data. Is the sales manager better off? That depends what’s being automated. If the underlying sales process is non-existent or weak, then automating it isn’t going to improve too much. It might be quicker to compile the data, but unless the process itself improves, then forecast accuracy is unlikely to increase, or the conversion rates improve.

The real benefits accrue when technology enables powerful processes. If those processes don’t exist already, or aren’t developed as part of the CRM deployment, then all CRM achieves is marginally more efficient disorder.

CRM technology selection...

The recent bankruptcies of Rover, Red Letter Days, and EU Jet, are a painful reminder that when companies go into administration the customer can get hurt, whether that’s the sudden depreciation of a recently purchased car, the un-experienced experience of a life time, or being marooned in a distant international airport.

Software companies of course don’t often go bust; they tend to get acquired because they have value in their customer base in the form of ongoing support revenues. These acquisitions can still be damaging. The acquiring company may have limited obligation to further develop or support a client’s software, or can charge punitive maintenance fees in the knowledge that it's rarely easy to up sticks and go elsewhere.

SME’s face a second potential point of failure in that they are often purchasing systems from Value Added Reseller’s (VAR’s). The VAR business model can be a precarious one, and companies not infrequently cease trading, or revise their portfolio of supported products. If the worst happens, clients can end up stranded with paid for support contracts which may no longer be honoured, uncompleted work, and undocumented customisations.

On the plus side, if one VAR leaves the business there are others to take its place. However there’s cost and pain involved in these transitions, particularly as clients are likely have gone through a run down in the quality of support and service as the VAR starts to struggle, or loses its focus.

With the CRM technology market in a major consolidation phase (Epiphany for example was acquired by SSA this week) and the economy looking less robust, wise technology, and if applicable, VAR selection, is going to be an increasingly key factor in getting long term value out of CRM.

Comparing CRM project day rates...

An entry in Joel Spolsky’s ‘Joel on Software’ blog caught my eye the other day. In ‘Hitting the High Notes’ he sets out some of the research he has undertaken, which seeks to establish the magnitude of differences in development productivity. His conclusions, which were well in line with other studies I’ve seen, suggested that there is between a 5:1 and 10:1 difference in relative productivity between programmers. In other words, what one programmer might achieve in an hour, another might not achieve in a day.

This set me thinking. Perhaps you’ve settled on XYZ CRM software for your project. The software is sold through Value Added Resellers (VAR’s), and because you want the best value for money, you ask a couple of VAR’s to quote. VAR A quotes a day rate of £750 per day, and VAR B quotes a day rate of £1,200. On the surface it looks cut and dried, VAR A seems to offer the best value for money.

But let’s say there’s a difference of 2:1 in productivity between the VAR B and the VAR A programmer, which is well within the norm from Joel Spolsky’s research, and the job would take the VAR B programmer 10 days. The cost for VAR B would be:

10 x £1,200 = £12,000

The programmer from VAR A, only being 50% as productive, takes 20 days to do the job. Therefore the cost from VAR A would be:

20 x £750 = £15,000

VAR B, despite the significantly higher day rate, is actually the cheaper option.

Which is all well and good, except that, even if you were to meet the developer assigned to your project, it probably wouldn’t be too easy, at least from a cursory glance, to determine their productive capability. The key to handling the variances in productivity, is to pay by the job rather than by the day. If say VAR A was prepared to do undertake the job for the sum of £7,500, the fact that would take them twice as long as VAR B is largely an irrelevance (bar perhaps the costs associated with a potentially longer lead team, and the concerns you might have around their financial longevity given the suicidal effective day rate) because they are now the cheaper option. You can only do this of course, if you spell out the requirements with sufficient detail and clarity that the vendors feel comfortable quoting on this basis. This being a to some extent a self serving blog, I will note in finishing off this post, that this is a service, we can, and are happy to provide.