Definition of an ERP project manager: someone who thinks nine women can deliver a baby in one month.
At Housing Technology’s annual conference back in 2013, I gave a presentation titled “E-RP?” to a packed room in which I restated the case for specialised packaged housing solutions (well, I would, wouldn’t I?). I followed this up with an article in this magazine, questioning the economic viability of the high-profile ERP projects in the sector. My argument was that despite the obvious problems presented by some of the established packaged housing management solutions, they do perform many essential tasks adequately which keep a housing business alive; rewriting this stuff in another technology was unlikely to improve things, and would ultimately be cost-prohibitive as a bespoke project or using top-end ERP platforms designed for richer markets. In short, don’t throw the baby out with the bath water.
Four years on, how has this panned out? It’s fair to say that any expectations that ERP would take over the housing sector, or that everyone would be using their finance system to manage rents and Dynamics CRM for everything else were wide of the mark (at least, so far). The high-profile ERP projects rumble on with mixed success, but there’s no sign that they will deliver solutions for the mid-market in the foreseeable future. Many Dynamics projects have delivered benefits, but to date no housing provider has managed to transition completely. Mistakes have inevitably been made, and, shock horror, it turns out that writing software is easy, but supporting it and keeping it moving forward are difficult and ever-more costly. Nonetheless, dissatisfaction with old-school housing management systems has grown and aspirations for a truly integrated and agile platform remain undimmed.
So if DIY and ERP are too expensive, how can a housing provider free itself from legacy, frozen IT systems to support the fast-changing business environment? Your core housing management modules are humming away in the background doing their jobs well enough, but you need to rapidly transform customer-facing services at the same time as extending your business operations. What’s needed, as has always been the case, is cost-effective and robust integration. ERP promises this by moving everything onto (apparently) the same platform, but this is simply swapping one monolith for another. In truth, it never has been technology that causes the problems; it’s the failure of suppliers to update it, play ball and generally do the right things. Being locked into SAP is no different to being locked into any of the traditional HMS platforms.
Warning: the next section gets a bit techy, but I will come back to ERP, honest.
If monolithic systems represent lock-in, then the emerging concept of ‘micro-services’ promises to break down the walls. There’s no official definition of the term, but the idea is to build applications from succinct, self-described and self-sufficient services which don’t rely on external business logic. Instead of having to deploy the entire application every time, each service can be deployed independently, enabling faster and less disruptive updates.
Some would say that this was the original intention of service-oriented architectures (SOA), and web services have been around for ever. Unfortunately, these often resemble the database ‘barfed’ into an API, and building logic into ESB or other middleware to cope with this just adds yet more agility-impeding baggage. By contrast, micro-services follow the paradigm of ”dumb pipes, smart endpoints”. They are designed from the outside-in, using a product-led approach which starts with the business need rather than with the existing capability. They enable a loosely-coupled architecture and are highly scalable, supported by a DevOps approach.
So what has this to do with ERP? The days of the ‘one platform to rule them all’ are coming to an end, and businesses now require solutions that can be quickly adopted with low time-to-value and implementation cost. If things don’t work out, they can be just as quickly dropped. The 2016 Enterprise Solutions Study by US-based ERP consultancy Mint Jutras surveyed over 500 businesses of all sizes and in various industries. Recommendations were to avoid “best fit” and instead look for “maximum agility”. You should also ensure that the platform is moving forward, integrates easily and can be customised easily without sacrificing support.
If software vendors truly embrace agility, and micro-services are arguably the best way to do this, then they will be able to respond to emerging market needs faster. They will be able to produce new solutions at a lower cost and with fewer dependencies. Furthermore, they can provide public APIs which can be much more easily used while still giving high flexibility. The risk to platform vendors is that they open themselves up to smaller but speedier and potentially disruptive competitors, but this is going to happen anyway; in the long run, it’s better to be ‘sticky’ and please your customer than to try to force them to stay with you by closing the shop. If you find your key suppliers getting in the way of integration, ask yourself what are the longer-term consequences of this, both for them and you?
Aidan Dunphy is head of product strategy at Orchard.