ashumskiy - Fotolia
In 1993, Tom Siebel launched Siebel Systems. Its flagship offering started off as a salesforce automation application; however, as Siebel saw the type of data that was being gathered, it became apparent that this could be used to gain insights into customers.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
As such, customer relationship management (CRM) was born.
Siebel Systems was by far and away the biggest proponent of CRM. But there were problems. CRM struggled to deal with the increasing number of channels (face-to-face, email, web, mobile, and so on) that customers were starting to use.
Others in the CRM space were facing the same issues, and CRM systems became over-complicated and too diffuse in their capabilities.
Also, the drive to gain extra information about prospects to provide more targeted offers was becoming intrusive.
Prospects were being driven away by firms that were trying to serve them better.
Turn CRM on its head
To serve prospects and customer better, CRM should be turned on its head. Rather than asking what CRM system should be acquired that enables the greatest amount of useful data to be generated, start from a position of: what data is already available?
The salesforce may be using Salesforce.com or a similar system. The marketing teams may be using a different system. The point-of sale systems in bricks-and-mortar environments may be using yet another. The website may be a
different one still.
Each of these is a data source that can help to create a full view of the prospect or customer. There are also external sources, such as Experian and Dun & Bradstreet, that can be used to add value to in-house data. Constituents of the overall value chain – suppliers and their sub-suppliers; customers and their own customers – may also have data sources that can add value.
Most organisations will already have systems in place to handle all this, and any advice based on a “throw it all out and start again” premise is likely to be met negatively. The presence of multiple existing systems must be accepted: the direction must be around getting the most from these systems, not how to jump on the next CRM bandwagon.
This is a data-first strategy. The first aspect of any “new” CRM approach should be how to get all these data sources together for analysis. The use of application programming interfaces and open data standards such as SQL make this easier, but not all sources provide such access.
Therefore, systems that enable automated data handling of sources where the structure may be more proprietary, such as Hitachi Data System’s Pentaho, will provide a simple and streamlined means of bringing data together. Indeed, a data-focused platform, such as from MapR, may be a good idea to deal with mixed data analytic engines such as Hadoop, Spark and Apache Drill.
There should also be a means of ensuring that data about a customer is complete. Too often, different systems operate using different key details to identify an individual. Is the field sales systems using “Jane Doe” as an identifier, while the marketing systems is using “Ms J Doe”? Is the logistics system using “123, High St, AB1 2CD” as a key identifier for the individual? A master data management (MDM) approach, such as provided by Informatica, IBM or Oracle, can create a single-entry system for users needing to analyse the data.
Read more about CRM
With all the choices of CRM for SMBs in today's market, picking the right one for your organization can be overwhelming. Here are some tips to help choose the one for you.
The right CRM platform can give your organization a big boost; not just in sales and marketing, but in every area of your customer engagement strategy.
MDM creates a single master record for an item – in the case of CRM, generally the customer. As long as entry to the underlying data is managed via the MDM system, any changes made to the customer’s data will be synchronised across all the underlying databases. This ensures that a single view of the customer is possible – something that is difficult to maintain if multiple discrete and disconnected databases are used.
If real-time customer data analysis is required, in-memory database technologies, such as Qlik, MicroStrategy, SAS or Tableau, are worth looking at. If your organisation is already using Oracle or SAP as the main “CRM” engine, then Oracle or SAP Hana in-memory systems may be the best bet.
Always consider the desired outcome for the business. If cross- and up-selling of items is the focus, then analysis of previous customers can show the “if <customer> bought <this>, they also bought <that>” scenarios. Make sure that systems are put in place that enable such offers to be made at the point of contact with the customer. It is a waste of time to follow up a face-to-face sales opportunity with an email some time later. Far better to ensure the salesperson has the opportunity put in front of them while the customer is there.
How about reducing customer churn? It makes sense to ensure that algorithms are in place to observe when an existing customer is likely to move away and which can provide them with reasons to stay, whether by pointing out what benefits they already have or through making a special offer to them through the right channel.
Attracting new customers
Attracting new customers? Ensuring that sales and marketing systems are fully linked, both at the data and the business process layers, can help to ensure that prospects are led through the right process to make it more likely they will get to see what they need to see.
Make sure that these are learning systems – too often, the likes of Amazon and other online systems make ridiculous offers well past the time the prospect has looked at a similar item (or even bought something) and moved on.
Make sure help is provided at all stages. Automated frequently asked question (FAQ) systems, such as that offered by Transversal, can help to provide direct information at the right time to the prospect or customer. Where needed, allow such systems to fall back to intelligent customer agent systems, with in-cloud skills routing and a capability for human agents to bring in other skills as necessary. Suppliers such as Interactive Intelligence, Avaya and Aspect offer systems for this. Other systems, such as eGain and Kana, allow automated multi-channel intelligence to be used in responding to customer requests via email, web chat, and so on.
Using both MDM and systems that can pull the data together makes a single pool for data analysis possible, even where that singularity is created around the use of virtual links between different physical databases. The move to more intuitive and easy-to-use data analysis tools, such as Qlik, Tableau and MicroStrategy, has further democratised the use of data analytics to those outside existing data scientist and analytics roles.
But there will still be a space for advanced analytics where the skills of the mythical data scientist may be required. Here, SAS, R, Hive, Shark and other tools can be used.
Embedded analytics engines are increasingly coming to the fore. A leader in this space is Logi Analytics, which has a specific focus on getting its highly flexible and visual engine into other software, such as with LSI, MotionSoft and Integral.
In line with the need to keep a focus on desired outcomes, don’t forget things like being able to plot data against geographic maps (such as via Esri UK, Google Maps or InstantAtlas) to derive additional data, such as time to get from one point to another, or points that are equidistant time-wise from a point (isochrones).
Analytics democratisation extends beyond your own employee base. If implemented well, it can be of use to prospects and customers themselves, allowing them, for example, to configure an item by size, colour and material themselves; as well as to the supply chain, where constituents can plan better for meeting just-in-time schedules.
All of this provides further data for your own organisation to analyse and use – without it being seen as intrusive to the customers themselves.
Data is key
So what does this all mean? It means that CRM as a monolithic, self-referential application is dying. It means that data is key and that organisations need to not only focus on what data they have internal to themselves, but also on other external data sources. The key to next-generation CRM success is in putting together a platform that is suitable for aggregating, analysing and reporting on the different data sources and types available.
It should not be a case of starting all over again, but making what is there work more effectively. And that will require a data-first strategy.
Clive Longbottom is founder of Quocirca.