Salesforce and the art of minimalism — part 3
And what actually is the MVP?
At its core, creating the minimum viable product means stripping out anything that is not strictly necessary for the function to work.
To illustrate this, it is easier to look at what the MVP isn’t.
How many times have you, as an admin or as a developer, been asked, ‘can you just?’ or ‘is it possible to?’.
‘Can you add a tick box?’ This is the classic ask and can be for literally anything. On paper, it is the ultimate MVP. Yet, without planning, adding a humble tick box, in isolation, is the easiest and quickest way to build up technical debt and create long term problems.
Other examples; ‘can you integrate this data enrichment app?’, ‘we want to implement account based marketing’, ‘is it possible to track competitors?’.
To each of these questions, the answer is ‘Yes’. They are closed questions that do not elicit dialog and, whether that’s the intention or not, it’s a problem.
Now ask yourself how many times you’ve been asked ‘how do we?’ I struggle to think of occasions when I’ve been asked to do something in this way. ‘What’s the best way to achieve this, Rob?’, ‘How can I achieve this result, Rob?’
Good question.
Really, really great question.
But, we’re always in a hurry and it takes time to find the optimum path. Often, we need quick results, to put a tick in a box (ironic) and get the job done.
Well it doesn’t work.
There is a saying in the audio production world, ‘rubbish in, rubbish out’. That’s not the original version of that saying, not the way I heard it anyway, but I’ve edited it because this article is safe for work.
The essence of that saying is that in sound production, the audio you record will never be better quality than when it was first recorded. The microphone, the cabling, the recording device and the environment; everything used contributes to the sound that is recorded and that recorded sound can only ever get worse. Think about that for a moment. Sure you can use audio processing to clean it up or whatever but it will be fake and in terms of pure data, it will only ever get degraded.
What does this have to do with making an app in Salesforce? The key word is data and the analogy tracks very closely to our topic. When you make the minimum viable product, you are focussing on creating the purest data possible. You are reducing your goals and metrics to the most fundamental elements and capturing only the required data in its most perfect format. The microphone, cable, recording device and environment in the audio recording are the custom fields, validation rules, screen flow and dashboard in your new app.
Once we’ve arranged and listened to our pure and perfect recording, we can get creative and add all the extra bells and whistles. We retain our original recording (data) but layer in the complexities, knowing that we are enhancing what we have rather than fixing it.
We perfected the basics first; we started with the MVP. Then we made our hit record; we built on a solid foundation to create the ideal product.
Start at the top and work down
Looking at the big picture is the essence of creating the MVP. Another contradiction? How can you consider everything and be minimalist at the same time? Our scope is growing, not getting minimised.
When we’re talking about creating an Org from scratch or reworking an entire object then yes, the MVP requires that you include everything so that you don’t have to make workarounds in the future. As we’ve already said, it means stripping everything out that isn’t strictly required. It means creating a simple, sustainable foundation on which you can safely build.
So you start at the top and work with what you know. You want a system that matches your customer journey. Surely that’s the best place to start? Yes, yes it is. Let’s work with an example of what a basic B2B SaaS based customer journey might look like. Let’s call the company Rob Inc. and that company sells an AI Assistant service.
Your customers begin their journey by finding your website after using a search engine. They sign up for a free trial of your software providing their name and email address to register their account. You now have all the data you need for your Contact record in Salesforce. You can safely delete, hide and remove access to the standard Fax field. Hoorah. They supplied their mobile number as their preferred method of MFA so it makes sense to store that data. The number is validated by a security check with the end user so we know it’s good. That validation process also triggers an automation that checks a ‘Mobile Number Validated’ checkbox on the Contact record, which cannot be edited by users.
Once entered and verified, the Mobile number should not be edited either so your data governance dictates that a validation rule should be written to secure this field. You could use a combination of the ‘Mobile Number Validated’ checkbox and a ‘NOT(ISBLANK(Mobile__c))’ check so once the mobile number is populated and validated, it can’t be changed.
So your entire Contact record consists of 5 bits of data so far; First Name, Last Name, Email, Mobile, Mobile Number Validated. That’s your MVP right there, all you need to make the system work according to the customer journey we’ve mapped to date. Let’s continue with the journey a bit more.
The free trial runs out and they decide they want to buy your software. The next customer touch point is a link to a web page for the end user to enter their payment details. So they provide their card details and their billing address which you store securely in a separate payment system. They agree to your terms and conditions and receive an activation code for 12 months system usage. You now have licence type, contract date and auto-renewal details to store in your system so you add the relevant fields etc.
The point I’m trying to make is that we only have a handful of fields to track and they are mostly locked down with very few users requiring more than read only access. This is the bare minimum required to make the system work and this is your MVP, your starting point. Sure you also want to add usage analytics so that you can identify upsell opportunities but this isn’t strictly necessary for the MVP we’re working on, so you can do that once you’ve got the basics in place and you know what? Those usage analytics are going to be implemented using the MVP methodology we’ve been discussing and it will be effective because it was built on a solid foundation of another MVP.
Key Data Points
How far can I take my sound production analogy? Let’s see. In music terms, the key data points are the notes that make the chorus, the harmony, the hook that goes round in your head and makes the song go to number 1. Yes!
What does this mean? We use systems to gather, generate, organise and present data. The ultimate goal of investing in a high-end CRM is to provide insight. Otherwise, you just have a database of customer details for processing orders. You don’t need Salesforce for that.
The MVP allows us to collect accurate, consistent and meaningful data from which we can derive true insight. This means identifying trends; refining your ICP, forecast reporting that facilitates Sales hitting targets month on month, identifying churn risk indicators early enough to reduce or prevent churn, highlighting the most successful marketing campaigns… The list is endless and every single function is enhanced by accurate, reliable data. Sure, you can get insights from any data but how much risk are you willing to take on a report that’s filtered on the wrong field and includes an amount field based on a formula that includes another field that’s based on a combination of another two formula fields that are populated by CPQ Summary Variables that reference a Cost Amount field that is no longer maintained because the users that configured and maintained those fields are no longer in the business and the details weren’t included in the handover… for example.
The scenario I’ve described above is not that far-fetched (speaking from personal experience) and really shines a light on how complex some data models can become. One typo in a cost field and suddenly the ARR GP formula is giving you an extra 0. You can close the books early this month, excellent.
Your key data points are the fields that are critical to the flow of data throughout your Lead to Cash cycle. The contact email address must be valid because it’s used in marketing, support and invoicing processes. The billing address, which includes the country that is used for marketing segmentation as well as Lead assignment and budget allocation. That one country field that is used in every area of the system, by every department. It has to be accurate because if its not then it requires 3 users in 3 different teams to agree what the country should be and correct all of the errors caused downstream because the country was wrong.
The time you saved by not planning in the early stages has a knock on effect that’s stopped 3 members of staff from doing productive, revenue generating work. Their flow has been interrupted and they’re spending their time compensating for flaws in the system. Imagine this happening on a day to day basis, every time a key data point is wrong then its someone’s job to fix the data. The impact can be widespread, immeasurable and costly.
Therefore, the benefits to mapping out your systems and the key data points are clear. Once you know the data points that are essential for your business to function smoothly, you know the data points to monitor and safeguard. You know that they need to be secured so that only the right users have access and you know that you want to protect them in your minimum viable product.
If you got this far then you’re really going to want to continue to the final thrilling instalment, part 4— The Big Finale… data governance
Missed parts 1 and 2? Here they are— Be the MVP, make the MVP and Planning, planning, planning…
