Best Next Action is a marketing technique when combined with sales and customer care provides better odds for selling and retaining customers.
The approach uses data mining and data analytics techniques to identify opportunities to apply to a person contacting the business.
This article discusses some of the fundamentals of the technique and how you can apply them in your business.
- Nothing new but very cool
- The Challenges
- BNA in operation
[read more=”Read more” less=”Read less”]
Nothing new but very cool
Definition and background
BNA (Best Next Action) is an area of marketing that is making the most of Big Data because large volumes of information facilitate the creation of models for what should happen next.
The approach is often used in contact centres because it can improve things on mass production scales, such as many agents taking contacts all day every day.
BNA can be defined as:
Next-best-action marketing (also known as best next action or next best activity), as a special case of next-best-action decision-making, is a customer-centric marketing paradigm that considers the different actions that can be taken for a specific customer and decides on the ‘best’ one. The Next Best Action (an offer, proposition, service, etc.) is determined by the customer’s interests and needs on the one hand, and the marketing organization’s business objectives, policies on the other. This is in sharp contrast to traditional marketing approaches that first create a proposition for a product or service and then attempt to find interested and eligible prospects for that proposition. This practice, direct marketing, typically automated in the form of a campaign management tool, is often product-centric, and usually always marketing-centric.
https://en.wikipedia.org/wiki/Next-best-action_marketing
There are mountains of material on this area.
Best Next Action in action
Two scenarios are very popular when BNA is called into action, sales and customer care scenarios.
Sales scenario
- Imagine a customer rings in to a sales person.
- What do they know about the customer? Nothing.
- A dialogue occurs and the sales person attempts to work out what is the best solution for them using a number of techniques.
Good sales people usually have a patter which they can use for scenarios but these are built up over time.
Customer Care scenario
- Image a customer rings in to a customer care agent.
- What do they know about the customer?
- The customer care agent should know an awful lot, but trying to collate all that information between the customer phoning in and the person answering the phone is where technology comes in.
Essentially Best Next Action uses statistics / models to guess what is most appropriate for that customer to provide an upsell or cross sell opportunity.
The more information you have the more tailored a response you can provide. Anything that helps to prompt the agent taking the call in saving the sale or proposing upsell / cross sell opportunities is good for the business. The systems are improving the agents delivery without the need to train the agents to spot patterns.
How it works
For both of the scenarios described
- The customer rings in.
- Technology through various means identifies the customer via their phone number, the number they dialed or through collected data (i.e. an IVR) or the whole thing detects a new customer not on record.
- The system based on which model the customer fits creates suggestions for the agent…
- These options are displayed on screen and allows the agent to naturally deliver the suggestion
A suggested Next Best Action is what is presented.
If you have no data on the customer, the agent is prompted to ask questions to quickly build up the profile to allow the system make its suggestions or learn from this new experience if it has never happened onto this type of customer before.
The challenges
Guidelines not rules
The dark arts of marketing and predicting what should and should not be done is not easy as a result of every customer being different with a different life story.
It is possible to make informed decisions with large data sets but the word to note is “informed” because Best Next Action is based on general rules from past experiences not guaranteed results.
Every case is unique and all of this work should be considered under the pretext of Best Next Action “is more what you call guidelines, than actual rules !”
Challenge 1 – The Data
Access to data is the first challenge and as a result if you want to build models you need good, up to date data. Garbage in, garbage out.
Collating all of that data is frequently a business’ first challenge.
Usually building models is done in a Data Warehouse where the data is clean and collated from across the business.
An emerging challenge with this area is the phrase GIGO which isn’t just an acronym for Garbage In, Garbage Out but also Garbage In, Gospel Out. Not understanding the how or why the models are the way they are results in misinterpretation of the results and consequently trust is place in the wrong conclusions.
Challenge 2 – Creating models / segments
The next phase is to create models within the data. So what does this mean? Well, image a very simple scenario. We have
- a date of birth field, which allows us to work out an age
- an address field, which allows us to group localities
- a firstname, which (in general) allows us to work out gender.
If they are an existing customer we know which products they have bought.
So
- Age
- Area
- Gender
- Product
From this simple set of data, we can use three different approaches.
Approach 1: The manual approach
This approach uses knowledge from sales people and experienced marketers who create “segments” consequently creating groupings of customers who are to be targets.
Approach 2: The Data Mining approach
What is data mining?
- “…the process of discovering meaningful new correlations, patterns, and trends by sifting through large amounts of data…” (Gartner Group)
- “…the analysis of observational data sets to find unsuspected relationships and to summarize data in novel ways…” (Hand et al.)
- “…is an interdisciplinary field bringing together techniques from machine learning, pattern recognition, statistics, databases, and visualization…” (Cabana et al.)
It is possible to elicit Knowledge in Data from statements such as “A particular customer shops in the Ballymount branch on Friday the 14 August at 9:15am, buying a box of tea and a litre of milk with a total purchase price of €3.10 using cash”
- A case is a specific collection of facts
- Facts are essential raw material for the elicitation of knowledge from data
- Knowledge is patterns or regularities identified in the data
- Patterns (regularities) in data that can be expressed by a statement of the form (IF x THEN y) are called Rules e.g. IF a customer purchases a box of tea THEN they will also purchase a litre of milk
This approach uses existing data and mathematical models to statistically generate models.
For each rule generated there is Support (how frequently these facts pops up in the data) and hence there is also Confidence (for all the times they show up, how often are they seen together)
There are a number of data mining algorithms available, each with different focus for model generation Prediction (Classification, Regression) and Description (Association rules discovery, Clustering). As a result of this this process you will have rules and consequently these rules are your segments.
Approach 3: The artificial intelligence / machine learning route
With modern computer systems capable of performing thousands of rules in real time, just think computer games consequently it is possible to build models that can adapt in real time. However they can be exceptionally expensive to build and are often very resource hungry.
It is worth mentioning that letting a computer make a decision on your behalf of your business without some level of control can be dangerous until the computer learns enough to behave in a manner appropriate to your business.
The way it adapts is to use the above data mining techniques in real time to see if the rules have changed. Artificial intelligence uses statistical tolerances which can change rules in real time when if finds better statistical results. As the models improve with more data they learn and consequently this is why the area is called machine learning.
The usual outcome
The usual outcome is a combination of all three methods.
- Start with a creative / manual process.
- Collect data and mine it.
- Refine the model.
- Artificial intelligence facilitates the up-keep of models and enables real time analysis.
It is also possible to use surveys and market research to add more facts.
Challenge 3 – Measuring results
Ok, so we have some segments we’re going to target. It always costs money to do marketing, so the most important thing to do is to create a Return on Investment model.
- What is our base line sales figure? (i.e. what do we make without any extra effort)
- What do we spend targeting this segment?
- When do we put the plan into action?
- The plan executed so what are the results? (The difference between now and the previous base line, is the ROI)
However being able to accurately associate marketing spend with outcomes requires a lot of systems, steps, checks and integration.
BNA in operation
Overview
Marketing drives communication into a contact point, say for example a customer care department.
Every piece of marketing has a unique code in it called a “media code”, to allow tracking however some methods are more automated than others.
The Best Next Action engine kicks in when the contact arrives and the engine presents the agent with its suggestion for a Best Next Action.
BNA Engine
We have all the key elements and as a result we just need to decide what mechanism is going to work best.
Scenario 1: Homework already done
Out of hours, a number of processes run which are the application of the segmentation and decisions based on the classifications above.
Essentially we predetermine based on the data we have what the best scenario to present it.
Pros
- Out of hours building helps when really complicated rules need generating.
- Reduces operational overhead of real time calculation
- Useful where systems do not have an API consequently batch processes help.
Cons
- Not easily adapted with live information captured from the customer
- The engine serves up the wrong rules due to inaccurate data
As the contact comes BNA simply looks up the result calculated last night instantly presents the BNA to the agent.
Scenario 2: Doing your homework on the way to class
As the call arrives, BNA looks up the key data and runs that data through a set of fast acting rules which calculate in real time and presents to the agent before the agent speaks to the customer.
Pros
- Rule sets change and update in real time
- Can have dynamically captured data added to the records
- Can read from multiple sources in real time.
Cons
- Not easily adapted with live information captured from the customer
- Increases operational overhead of real time calculation
- The engine applies the wrong rules due to inaccurate data
- Lookups / copies of data must be available for real time interrogation
Scenario 3: Doing your homework in class
Best Next Action calculates as the agent is able to ask the customer questions as a result the agent gets the most dynamic solution of all consequently compensating for lack of data. Real time data runs through the rule sets and BNA calculates live but you need your systems to respond in a timely fashion.
Pros
- Rule sets change in real time
- Dynamically captured data added to the records
- Can read from multiple sources in real time.
- Live data can make sure Best Next Action is best available with live information
- Adapts for bad / out of date data sets.
Cons
- Not many!
- Increases operational overhead of real time calculation
If you would like to discuss any part of this article please get in touch here or on social media.
[/read]
