Analytics and workflow
There’s a big difference between B2C and B2B analytics that no vendors seem to be addressing and it involves the consumption model. I spoke with K.V. Rao, founder and chief strategy officer of Aviso, an analytics company focused on sales recently and his unabashed opinion is that, “If you’re trying to expose insights, and make things consumable, you have to address workflow.”
He has a very good point, especially for buyers who may be having trouble figuring out what they need. The decision process runs trough digital disruption (am I being left behind?) to big data (what do I do with it?) to analytics and machine learning (same stuff, right?). Usually at this point in a client discussion I ask people to tell me the kind of information they want to get but now I think this might be jumping the gun. Before asking what kind of information you want, it would be an excellent thing to better understand what processes or workflows you’re trying to influence.
For example, we all want to sell more and we’ll try almost anything to do it, but that opens a big can of anacondas. What part of the marketing and sales process do you want to influence or spiff up? Is the problem quantity or quantity of leads? Do your reps get stuck in one part of the sales cycle? Are renewals off? Do your customers up and leave without warning? We could go on and there are analytics tools that can help with all of that but you need to get the diagnosis right.
Workflow may not be the first thing that comes to mind when considering analytics but it might be the big silverback gorilla sitting in the corner. Workflows are vastly different in the B2C world than in a B2B situation. Simply put, when dealing with consumers, analytics is aimed squarely at aiding customer decision-making in the moment, so as Rao points out, “The workflow is almost non-existent.” Good point and not at all what decision-makers in the enterprise encounter.
Consumers are trying to figure out whether or not to buy and that’s rather binary. On the other hand, enterprises buy by committee and need to develop information from whatever data they collect so their need is for long term information to build a purchase case.
Marketing, sales, service and support might all benefit individual users through analytics for in-the-moment tactical decision-making but only to the point that they are already working on organizational goals and not individual choices. Last week IBM and Salesforce introduced a new partnership based in part on their respective analytics tools, Watson and Einstein. There’s great interest in how these two will work productively together without stepping on the other’s toes and the evolution of this relationship will say much about the future of analytics generally. As I see it and as the press release strongly hints, Watson will be important for providing strategic situational information while in CRM at least Einstein will support the tactics of a vendor-customer interaction.
One example in the press release was Watson providing retailers with weather forecasting information that could easily be applied to better understanding the traffic pattern to expect for the day. Einstein on the other hand would still be responsible for understanding data about customers past purchases, new requirements, upsell and cross sell potential and more.
Retailers have had all of this in mind for a very long time even before Watson and Einstein and they coped though not always well. As Mark Twain is supposed to have quipped, “Everyone complains about the weather, but no one does anything about it.” And as the great retailer John Wannamaker or possibly Marshall Fields once supposedly said, “Half of my marketing budget is wasted. I just don’t know which half.”
So the different roles of analytics are intended to solve those and similar problems but before they do we still need to get a handle on the workflow we’re trying to influence. That’s still a job for the human mind as is the most important decision of all—whether to accept an analytics driven recommendation or to make a decision to rely on other information that the genius software is not privy to. For instance even with a great weather forecast it’s probably still vitally important to know if there’s a parade coming through downtown at noon.