Automating For The Process?
What Makes A Good Candidate For A Process To Automate?
If its repetitive and open to human error.
People are actually very good at tasks with a lot of variability but lets face it our concentration starts to wander if we have to do the same thing over and over again.
So;
Tasks that happen every day and follow a clearly defined routine or plan are ideal for automation.
Tasks that you do once in a blue moon are not great candidates.
Think about how long it will take to automate the task and how much it will cost. If the cost of automation is greater than any benefit you will derive there is little point in automation. To do that you need a good understanding of the costs associated with the process and the cost to automate it.
To be able to make that sort of judgement you really need to understand how your business works. Mapping out the process flow around an organisation is key to that understanding. Building a customer journey or just a plain old flow chart is a good place to start.
Then you need to critically evaluate that process. There is no point in automating a process that is fundamentally inefficient; all you are doing is building an automated version of the inefficient process.
You can get to the stage where you know on paper what your process ideally would look like. Now you can have a conversation about how to automate some or all of it.
Having an end goal is really important; to improve efficiency, enhance customer service, reduce costs, or scale your business that sort of thing. Its important to have a means to measure your progress against those goals though. Ideally you will break your automation into manageable steps that make it easier to measure progress against.
This is often referred to as the Discovery phase of the process.
How Do I Find The Best Way To Automate My Process ?
Once you have a process that you want to automate you can move onto what is probably best referred to as the shopping stage. (though is probably technically just Design)
You need to find the tool that you want to use to perform the automation or someone to write the software for that tool.
The important point here is to bear one thing in mind.
Most businesses are the same bar a relatively small percentage of their activity that makes them unique. That’s how Software as a Service companies like Salesforce and Hubspot got to the scale they have. They have been able to abstract CRM related processes so that one piece of software can run the same business activity in a range of business types.
As an example; if you think about retail; on the face of it B&Q looks very different to W.H. Smith (beyond just having bigger orange and black stores) but if you went and looked at their back office functions they will talk similar language; stock, products, shops, staff, sales, refunds, stock control. The differences are in the physical nature of the product. But if you think about the businesses in the abstract sense they should have very similar processes and you can measure their process using similar types of KPI.
That’s obviously if you exclude the historical factors that lead to us having a convoluted set of business processes but going back to the Discovery stage these are the things we want to be taking the opportunity to design out.
In theory you should be able to take a back office system that supports B&Q and apply it to W.H. Smith with a small amount of customisation for the outlying business cases.
If you are running a retail business and your process looks significantly different to what the off the shelf software packages you are offered will support you need to take a long hard look at yourself and ask the question why is my business so special ?
I’ve Bought A Bright And Shinny New System Now What Do I Do ?
This is the Build/Configuration/Implementation stage.
Rule one .. Start small, really small.
Your team needs to be able to absorb any change you implement. Starting on a non-essential low stress part of the business will give them the opportunity to learn how the software works. The rationale and language it uses can be understood and gradually brought into the business on a day to day basis. Pick the right people to be part of this path finding group. They need to understand your business and have credibility within it. It also helps if they can understand IT and explain reasonably complex ideas well to other people. They will have to answer a lot of questions.
Given the comments above, that you are not that different to all the other businesses in your market sector, there will be new terminology to learn; it may be a question of discarding language that has been used for years in favour of a more industry standard approach.
There is a temptation to rename all the fields on a new system to match the names you give things in the business; but realistically if you do that you are introducing something to maintain going forward. Longer term its better to use terminology that is industry standard where it exists. That has the added benefit that suppliers and customers will better understand what you are talking about.
Avoid massive amounts of customisation. Customisation builds in maintenance costs going forward.
If you need a lot of customisation to make the system work you need to consider two possibilities;
You picked the wrong bright and shinny piece of software; it shouldn’t be an issue to switch horses at this point in the race if you stuck to the “Start small, really small” mantra.
You weren’t honest with yourself at the Discovery stage.
Multiple stages are best rather than implementing everything at once. I always think 3 month stages work well; long enough to implement something meaningful but not so long that you to lose sight of the stage goal. This minimises disruption and allows you to fine-tune the process based on the what you learn at each stage.
And finally don’t be afraid to admit you picked the wrong horse. It happens. The key is to learn from the experience, you should be able to use the material from the Discovery phase to progress with a different approach without having to go back to square one.
Be honest with yourself around why the first implementation failed.
So What Do I Get Out Of Automation?
Lots of things.
Less double keying; Double keying or typing the same information out multiple times into different systems is a huge time waster. Its also a great source of inaccurate data. Think about it; who is the best person to correctly type the customers address; your admin staff or the customer. Get the customer journey right and the customer and your new automation can do a lot of the heavy lifting for you.
Consistency of process; You will have a more consistent process and customers will have a better idea what to expect. Automation can ensure a more consistent standard of service delivery by reducing the risk of human error. Automated workflows follow predefined processes, reducing the likelihood of mistakes, and increasing customer satisfaction through the delivery of high-quality experiences to clients.
Notifications, lots of notifications; We all like to get notifications when our Amazon order is dispatched. It would be really tedious(and expensive) if a member of staff was typing the email each time but if you can automate the process its relatively painless from a business process viewpoint.
When staff dispatch the parcel the message can be generated automatically. Each time the parcel moves another stage through the fulfillment process it can trigger another notification to the customer. You want to give the customer the opportunity to opt of those notifications but that opt out can also be automated. “Don’t want these messages any more … just follow this link madam”
One word of caution I would add to this. Beware of unintended consequences.
The customers will for instance know where the parcel is all the time and they will know when it sits in the customs shed at Heathrow for a week not moving. You need to be prepared for the “why hasn’t my parcel moved questions”.
Data lots of data; Automation has another happy bi-product. It gives us lots of structured information that we can use to improve the efficiency of our business. We are recording the progress of items through the business so we know where they all are, how many we have, how long it takes for Fred and George to send the parcels, Fred and Georges productivity and so forth. You have information that can massively improve your business decision making process.
Scalability; Automation also gives us the ability to scale. As your business grows, managing an increasing volume of tasks manually can become overwhelming. Automation provides a scalable solution, allowing you to handle a larger workload without significantly increasing your overheads.. apart obviously from paying for the computer resources; beware AWS Credits .. at some point you will have to pay for your computer time and its best to understand how much it will be at the beginning.
You should be able to achieve that scale without significantly increasing your overheads. The automation should scale without the need to hire additional staff; certainly in the first instance.
This is particularly significant if you are automating activities that don’t necessarily generate sales. So for instance automating the process of producing quotes. If you can setup your website so the majority of the people your call centre talks to are ready to purchase rather than seeking further product information you will have a much more scalable system.
Think about quoting for household insurance. We all shop around when we look to renew; without automation we have to talk to someone to get the price and the details of the cover. We are shopping around so in most cases we won’t actually make a purchase at the point of initial contact with the insurer. Automation of that part of the process makes a lot of sense. Bear in mind though that your automation has to work, having a badly performing quoting system is like having a shop door that is jammed shut.
Smoothing out the ups and downs; Automation also allows you to scale back the other way without necessarily impacting your workforce. In businesses with a cyclical business cycle being able to cheaply scale down in the down part of the cycle is very attractive.
Remember; regularly monitor the performance of your chosen project and analyse how well it’s working for you. The change in productivity should be noticeable. Once you are satisfied that the project has increased your efficiency, then you can start to think about what to automate next.
Want to talk about process automation? Click the button to book a call with Be Astute.