Blame it on the Process

Process. Process. Process. I remember hearing my senior colleagues saying these words in the near past and yawning.

Who cares about process? What even is a process? Why should I care about the process -I just want to do cool things!

I was a little naive. Process is everything, and an integral part of design. Understanding the requirements of a flow are vital when it comes to designing a process it can follow - think roads, cars, buildings, logistics - and of course data. Design right, people will flock to it - design wrong and they’ll soon let you know about it.

In many cases the high level process is fine, it’s logical and often there’s little room for change at that top layer, for example:

Gather Data > Modify Report > Get customer approval > Publish Report

It’s a simple process, and it works - to publish a report this process is required. It’s worked for years, so why even think to change it? Well let’s break down each step of the process and see what effort is required to accomplish them.

Gathering Data

This is very vague. Where is this data coming from? A single source of truth, an excel(s), a person, a report, the sky, all of the above? Before I talk about what I envision as a best-practice let’s think of the reality. We’re publishing a monthly performance report, we have a database but the data isn’t always accurate and isn’t always up to date. This means the data we pull from here often has to be modified by stakeholders who know the actuals. To do this we send emails, we have phone calls we fire around excels. Once we’ve accomplished this we consolidate all of that data, and sned it around to our stakeholders again to make sure we’ve not left anything out. It’s not fast, simple or transparent but eventually we get what we need - now we need to move on to step two.

Update/Generate The Report

Again, let’s talk real world. We’ve not got access to a fancy analytics tool, connected to a Single Source of Truth. So now we must look at that consolidated data - and format this into a report. We may have a template, or a macro to do this for us, or perhaps we have to do this click-by-click. Whatever method we get a report. We may have to compare this to the previous one with the help of some vlookups - and then we’re probably going to have to show the stakeholders the report again, to make sure we’ve not missed anything out. Again, no speed, no simplicity - but we get there in the end. We have a report - let’s get our customers approval.

Approval

Now we hope that our customer is filled with glee at our fancy charts and colourful spreadsheets but, they aren’t. Something’s not right. People in the room are pointing to figures and asking questions. I mean, it’s been proof-read several times but, y’know things slip the net don’t they. Only… your customer isn’t going to take that for an answer, and guess what - they want to know exactly where that figure came from. $%#*. You see, the thing is, we don’t know - there’s a process guidance document somewhere which might have the answer, but the report has been handled now by so many people and there’s so many versions - we’ve lost all traceability. This means we don’t know who is accountable. This means we’ve got to do a lot of searching to provide our customer the right answer, if any answer at all. It may also mean after all of that hard work - your customer will dictate to you exactly how the report should look.

Our process at it’s lowest level has failed. So, how can we stop this from happening again?

Control your Data

We need to own our data:

Plan A. In an ideal world we’d tackle the very database the majority of our report data comes from - and make it a single source of truth, this is the best thing to do. This may mean a total overhaul, or some minor tweeks to the system. Once complete we’d have a single source of truth, that all of the data we need is fed into. It’s always up to date, always reliable and totally transparent.

Plan B. We may not own that database, or be able to change it. It may not even exist. This doesn't mean we can’t own our data. We can create a single source of truth for our report(s), it’s not the leanest method as it will duplicate data from the original source, but it’s leaner that what we had. This source should always be accessible to everyone, giving our stakeholders access to it so they can add their information at will.

When done correctly, these two steps will eradicate the need for email chains, multiple excels, 5pm panic calls. We can see the data, they can see the data. We can set some simple rules such as “Person X must update data by Y” and maybe even trigger some automatic notifications to keep people in the loop, establishing clear accountabilities.

Let the computer do the maths

Now we have our Single Source of Truth, we shouldn't concern ourselves too much with manually entering that data into a spreadsheet. We should be building a report, connected to our database so that we can see in real time what our data looks like. From personal experience this step can take an incredibly long time to get right, but when it’s done the benefits are immediate. People will be amazed that they can at the push of a button see a report instantly. No more waiting for you to run a vlookup and down an espresso - it’s there. In front of them. This should help errors be spotted before they even get to your customer - oh and guess what, now you’ve established accountability and transparency the person who owns that data will probably fix it, with little to no encouragement.

Now we have our report we can get approval. We can provide an answer immediately to the dreaded “where is this from” because it’s in our Single Source of Truth. If amendments do have to be made, we can make them in our database, we’ve got traceability. Our customers can be confident in our data, they are seeing the benefits our new automated process has - and by adding a feedback loop into the mix we can begin to make tweaks to our process design, so that at the end of each process cycle our customer sees change - and guess what, implementing it was easy.  

It can be all too easy to sit back and blame the database for being too complex, or staff for being too slow… but we’re now living in a digital age where we have the tools (often for free, or for little cost) to make these changes happen. It could be a specialist tool connected to a database, it could be a Sharepoint list linked to an excel - or it could be as simple as an online data sheet everyone has access to. By simplifying the steps in a process, and removing as much human element as possible we can generate results faster, act sooner and work smarter.

Simple process automation can lead truly positive and value adding transformation.