Top 5 Salesforce Sales Cloud Implementation Mistakes

Making mistakes is human — we make plenty of them every single day: we speak the wrong way to someone close to us, we forget to put on the parking brake after leaving the car, or we add three teaspoons too much salt to our lunch. But what if some mistakes are repeated? What if we see them over and over in certain people or behaviors? Wouldn’t it be better to avoid them? That’s exactly what this article is about — avoiding mistakes during Sales Cloud implementations.

I’ve been working as a consultant/solution architect for quite a few years now. I started with small projects, moving through various business sectors (recycling, loans, non-profit organizations, and many, many others). And I ended up working on large ones, where these mistakes also occurred. In each of them, I could find a lot of elements that needed improvement. And I’m not pointing fingers at who made those mistakes — because I made them too.

In 2004, a great movie called The Butterfly Effect came out, illustrating the results of the theory with the same name. The theory itself says that “the flap of a butterfly’s wings on one end of the world can cause a tsunami on the other.” It’s exactly the same with the mistakes made during the implementation of any solution. If we don’t follow so-called “best practices” or proven methodologies, we may encounter future issues in our work — incorrect report results or unrealistic sales outcomes that later turn out to be a financial mirage.

Here are 5 mistakes we may encounter during a Sales Cloud implementation:

1. Neglected data

"Please upload the data, we’ll clean it up once the system is ready" — I’ve heard this sentence so many times. And interestingly, when visiting the Org during later improvements, I saw the same data that was uploaded a year earlier. Full of “John QWERTY” and “Jane ASDF”.

During the implementation of a given solution in Salesforce, it’s important to remember that data is the battery that powers the system. It’s the clean and accurate data that provides users with reliable results on their favorite sales dashboard. Or forecasting — if we have garbage data in our database, we can expect outcomes that will either bury a salesperson or make them look like a fake millionaire.

Implementation teams focus on using modern solutions, networks of integrations with various systems, and confetti when closing a won opportunity (though honestly, I’ve never met a client who actually wanted that set up), and forget about one of the most important elements — data cleanliness.

But what does this lofty phrase really mean — “data cleanliness” — and what can actually threaten it?

• Duplicates. A common pattern is: “I forgot this client was already in the system,” and the person enters them into the database again. Let’s say we go ahead and upload such data without cleaning it first, and it turns out that thirteen contacts with the same first name, last name, and email address together have 26 opportunities. During data upload, the duplication rule can behave in different ways depending on how it’s configured — and we can either drive the admin crazy by throwing the same error again and again during the upload, or we might temporarily turn the rule off just to have some peace and quiet.
• Outdated contact information. Now that’s a tough topic. Why? Because it’s difficult for us to verify. This is mainly the responsibility of the person requesting the upload. And in this case, it’s not just an issue with fresh implementations, but also with long-term system administration. Make sure the client verifies the data, because it’ll come back to haunt them when they try to reach out to a customer and none of the contact information points to anyone.
• Lack of data entry standards. Oh, that’s a nightmare. I don’t know if you’ve dealt with this before – for example, when a field that was supposed to contain an email had either “alex@customcompany.com or maybe alexander@customcompany.com.”(Yap, both addresses with “or maybe” in the one line). And thanks to these little surprises, Data Loader flipped us the middle finger and said, “Sorry, your data is invalid.” The same thing happened with other data, like phone numbers. Why did this happen? Because there was no validation, no training for the team on the company’s data standards, or simply because there was one anarchist who thought, “Anarchy in SF!” and entered the data however they wanted.

And how can we fix this? Is there any way to clean this data? Yes! But remember — it should be the data owner who takes care of it. That person knows the data best. But how can you help prevent it? With four simple rules:

• Run a data audit before migration. Review the data and see what’s in there. Go through it in the context of actual processes. Take a look to see if there are duplicates or incorrectly entered records.
• Ask the client to deduplicate the data and remove any dummy data.
• Do a test upload. Verify the data you’ve uploaded (you can easily delete it later using the Mass Delete Records function). Suggest necessary changes to the client.
• In the system, design processes are used to prevent duplicates from being created.

2. Lack of proper training

Sometimes it’s hard for us to remember that there are people who don’t know what Salesforce is, and “Account,” “Opportunity,” and “Validation Rule” sound to them like spells from the Harry Potter universe. I’ve caught myself so many times, getting into the flow during a conversation with a client and bombarding them like a machine gun with terminology straight from the depths of Salesforce.

We can never assume the client knows what SF thing means. If we’re implementing a solution on a fresh org and our main stakeholders come from departments where spreadsheets were the primary tool — then hold your horses and start talking as if you were explaining something to a child. The client will thank you for it.

What can the lack of proper training lead to?

• Low adoption rates
• Reverting to old habits (spreadsheets, sticky note reminders)
• Incomplete data in the system
• Frustration within the team

“But wait, we gave them a great system! It does so many things, calculates values, updates fields, sends reminders — why aren’t they using it?” – Because they simply don’t know how to use it.

Let’s remember that for new users, using Salesforce is like us learning how to fly a plane (I was just recently being taught how to plan a flight — now that is real rocket science). A user has to step out of their comfort zone — which is spreadsheets and notes — and then befriend the new system. And it’s our job to help them do that.

Here’s how we can make it easier for them.

• Create a training plan tailored to specific roles. Don’t teach accountants how to manage opportunities, and don’t teach salespeople how to create new users. Adapt the knowledge to their needs. Prepare real-life scenarios based on their roles in the company.
• Appoint user champions. They’ll be the ones to spread the truth about SF… well, more seriously, they’ll take on the role of trainer. Pass your knowledge to them so they can train new users themselves.
• Prepare support materials and share them with users. If the existing trailmixes aren’t enough, create your own. Prepare a step-by-step presentation on how to work in SF — and most importantly, record your training and share it with users or place it somewhere easily accessible. Why explain the same thing three times if you can do it once?

Remember — even the best-configured system won’t bring any value if users are afraid to use it.

3. Improper segmentation strategy

Segmentation — that is, placing someone in a specific sector. Without proper alignment of marketing materials, communication, or customer interaction, the sales team won’t function effectively. Imagine, for example, that you’re selling PS5 games but your communication is aimed at hardcore PC fans. They’ll eat you for breakfast and hit you with a #PCMasterRace reply. Segment wisely — otherwise, you’ll face these main problems:

• Treating all contacts the same
• Lack of personalized communication
• Ignoring the customer’s previous interactions

Come on — we now have Agentforce on board, which helps us create personalized content. And if we convince leadership to go for Data Cloud, then we can really group people properly.

I’ll end this point with a reflection: Do you know your customers well enough to deliver valuable and relevant messages to them?

4. Underestimating implementation requirements

And this one’s a deep topic — sit down, grab some popcorn, and let’s begin.

Okay, seriously though — I think we’ve all been there: Sales promised something, and then you, as the consultant, developer, or architect, had to seriously hustle to meet the client’s expectations and stay within budget. I’ve been in that exact situation. My personal favorite? When a client asked for a Lamborghini during discovery workshops, and what we had for them was a polished, second-hand Opel Corsa (I mean in the budget terms).

If you weren’t involved in the project’s presales phase, then before implementation, go over the Scope of Work vs. Budget thoroughly. It’s critical.

Typical issues here include:

• Time required for process analysis. We’re not experts in everything — no one will fully understand the construction industry, business processes in a chain of garden centers, and the approval matrix in the metallurgy sector in five minutes.
• Budget for customizations and integrations. “But there’s a connector on AppExchange…” — that’s the kind of reply you might get when you ask, “Why are there only 2 hours allocated for accounting system integration?”
• Data migration complexity. Yes, this ties back to the first point. But it’s worth adding that if the migration involves Account, Contact, Lead, Opportunity, Product, Pricebooks, and Files — well, you’re going to be in for some fun with that data.

So what can we do about it?

• Conduct a thorough needs analysis
• Plan a realistic timeline with time buffers
• Work with an experienced team familiar with Salesforce

When it comes to implementing a CRM, you can’t rush. CRM is not just a tech change — it’s a digital renaissance for the company: optimizing processes, breaking bad work habits, and improving current operations.

We’re stepping into the 21st century.

5. Lack of business engagement

At the beginning, it’s worth asking the question — who implements Sales Cloud? A Salesforce partner or the client?

If you answered “the Salesforce partner,” then sorry, but you’re wrong. The client also implements the new solution. Without input from the business side, we won’t know what’s actually needed. It’s a bit like buying shoes — if you walk in and say, “I’d like some shoes,” you’ll get a default model, even though what you really dream of are Vans. But if you say, “I skateboard and I need legendary, durable shoes,” then among a few brands, you’ll find your Vans.

It’s exactly the same during a Sales Cloud implementation — the client tells you what they want. You, as the consultant, advise and encourage them to take advantage of certain improvements.

What we often see during implementations:

• The project is treated as just another IT initiative
• No business owners for key processes
• No clearly defined business goals
• Absence of business leaders at critical meetings

A quick comment here:
Everyone attending the workshops needs to understand that this isn’t just another small IT project. They need to know that this is about digitizing their work, improving their daily tasks, and — most importantly — making their lives easier. And even though we’re doing it digitally — come on — we’re changing their lives!

If we’re making changes to accounting, sales, or complaint-handling processes, each of those processes needs an owner. These owners must take responsibility for the changes introduced. Goals need to be defined at the very beginning — without them, we don’t know what direction we’re building toward. And the absence of key leaders shows a lack of commitment from the project’s main sponsors. It’s a bit like a class without a teacher — everyone does what they want, and no one answers questions.

But this can be fixed:

• Involve business leaders from the very start
• Assign business owners for every major process
• Define clear goals and success metrics
• Create a mixed team of IT and business representatives

If you meet these requirements, life will be easier — for you and everyone involved.

Summary

Are these the only 5 mistakes? Probably not.

Every consultant, architect, or developer has their own “favorite” mistakes. The ones I’ve listed here are, in my opinion, the most common ones I’ve seen in my implementations. If we avoid them, our future projects will definitely be smoother, faster, and — most likely — less patience-draining. And even if we eliminate these 5 mistakes, new ones will surely surprise us — and our stories at industry meetups will be full of them.