As a consultant, I’ve worked with companies of all sizes, ranging from a five-person startup to a few Fortune 100 conglomerates. Young companies have the benefit of flexibility and speed. They’re agile and scrappy, making do with what they can afford. Larger companies have more processes and tend to be painfully slow to make a decision.
Each stage also has their own systems quirks beyond speed-to-change.
Startups are more prone to switching CRMs and under-hiring for administrator roles, creating a lot of technical debt to address once (or if) they scale. Growth companies are busy trying to support changes in their go-to-market strategy while paying for the sins of past configuration decisions. Enterprise organizations can be more security conscious and want to lock everything down, can rely more on external contractors, and they’re also more prone to merging multiple CRM instances.
We’ve organized lessons learned by maturity and pulled quotes from community members as they reminisce about their own faux pas or grumble about what they’ve inherited.
Feel free to skip to the section that most applies to your situation:
Trevor Greyson, Head of Business Technology at Miro explained the Startup CRM phase like this:
“A company’s startup phase is similar to creating a foundation for a building. If you prioritize low spend (particularly around talent) and speed over proficiency, you’ll negatively impact your ability to scale. If you use bad construction materials, you may quickly have a building two or three stories high, but anything taller than that will be too unstable. You’ll have to tear it down and start over again.”
Let’s look at some lessons learned from the community.
Chances are you won’t get the technology you want out of the gate. Most companies start with a freemium platform (or even Google Sheets) until they’ve raised sufficient funds.
If you start with a free platform, think through the data structure you want for your next CRM or Marketing Automation Platform. Learn about the standard objects, research plug-ins, and limit the number of fields you collect.
Just remember, the more complicated a CRM is to use, the less often people will use it. Consider using conditional page layouts, object processes, and regularly purging fields that are no longer maintained to make what your end-users want to know about a company, person, or deal more accessible.
When you get the funding to upgrade to “real” SaaS products, hire an expert to implement them.
Do not, under any circumstances, let your freshly minted Salesforce admin run with the initial implementation.
You may think you’re saving money, but I’ve worked with too many organizations that have decided their only option was to rip and replace their original instance. A bad data structure, reliance on too many custom objects, a proliferation of bolt-on tools, and lack of unity on data definitions (aka horrible data hygiene) can snowball into a big mess.
Matthew Volm, CEO and Co-Founder of Funl, says:
"I know one big thing that I saw value in from my time at Ally.io was implementing SFDC from the beginning and hiring an SFDC admin consultant to help with setup and configuration to ensure we got up and running correctly. The company was in a fortunate position to do this since they were successful in fundraising efforts, but this laid a really strong foundation."
When you’re considering a consultant, insist on speaking with prior clients, and if you can, use someone you trust with the right expertise to screen them. It’s worth shelling out more money for someone who truly knows what they’re doing. If you have a bad feeling about someone, don’t hire them.
The longer you work in revenue operations, the more skeptical you’ll become of claims made by SaaS vendors. The only things that are certain in life are RevOps technology skepticism, death, and taxes.
I, too, started out naive and hopeful. Who doesn’t want a silver bullet that actually fixes your problem? Before long, you’ll be strung along by enough companies peddling vaporware that every vendor is doomed to intense scrutiny.
Now, I can’t prove that sales and marketing departments are particularly susceptible to strong marketing tactics, but I’ve had to talk enough leadership teams out of handing over the corporate credit card to know that you have to watch them like a hawk. They’re like kids discovering sugar for the first time, and sometimes (usually) they don’t know when to walk away.
If the solution sounds too good to be true, it probably is. Get the department’s requirements and insist on attending a sales call to ask qualifying questions.
No one wants to look like they don’t know what they’re doing. To avoid everyone throwing the initials “MQL” around to describe everything from an unknown prospect to a qualified opportunity (true story), work with your business leaders to define key metrics as early as possible.
Research best practices, join a Slack channel or online forum to get advice, and talk to your network. Reading blogs like this one is a great idea, too (fun fact: you should bookmark us). Don’t go into a meeting with leadership without being armed with enough information to gently guide them in a good direction.
If you think data is boring, you’re probably not in revenue operations. We tend to see value in order and process, and we like to know that the data we’re pulling for key reports isn’t a mess. This requires some thought to go into naming conventions.
Lance Thompson, Technical Sales Operations Manager at SeekOut says:
“It’s essential to get your framework in place for the reporting you want today and the reporting you know your leadership team will want in the future."
As an example, thinking through the ways you will need to organize marketing campaigns two or five years from now is very smart. Assume you’ll dabble in the usual channels, think about variations or sub-classifications you’ll need to support more detailed reporting, and anticipate the demand for some flexibility.
The only constants are change and RevOps technology skepticism.
Striking a balance between realism and negativity is a delicate dance. I’ve yet to meet a leader who appreciates being repeatedly told that a request is impossible. They prefer solutions, not problems.
But.
Not voicing a system limitation is setting yourself up for an even more disappointed executive team.
Recently, I helped a company assess its marketing reporting options within ZohoCRM. Having never worked with the platform before, I was unprepared for how misaligned their technical documentation was with reality.
They had screenshots of the system doing a thing, but the system absolutely does not do that thing in practice. Then begins a several-week waiting game bouncing around technical support tiers only to finally be told that the system is broken ("Thanks for bringing this to our attention!"), and your request is now on the developer's list of things to do. Some day. In the distant future.
We had to have a difficult conversation with the executive team. Campaigns were not associated with opportunities. They weren’t even associated with contacts and leads in a related object. They showed up on the screen but weren’t available in reports (it still blows my mind 🤯 ). This meant opportunity source reporting, funnel reporting, and marketing attribution reports were impossible with the technology they had in place.
Ultimately, they made the decision to stick with the tool and rely on proxy reporting (website analytics, form fills, online event registrations, and other disconnected reporting that is not integrated with the “source of truth” system) rather than invest in a more expensive solution.
Early in my career, I worked for a brilliant boss who hammered home that I was an advocate for people in the field. He forced each of us to participate in sales calls, ride alongs, and major meetings. We witnessed the stress of juggling a career on the road with an aggressive quota.
Executives frequently insist on data to help them make decisions. From their perspective, it’s a no brainer. They need a metric to figure out the source of an issue. Their employees are paid a salary, and data entry for the sake of selling more product is certainly part of fulfilling their job description.
Unfortunately, it’s never as simple as requiring one piece of information.
Lance Thompson, Technical Sales Operations Manager at SeekOut said,
“I’ve seen salespeople feel like they aren’t a valued member of the company because of the volume of fields they have to fill out ‘before they can do their job.’ It’s not just the one request for data--it’s the holistic setup.”
Look beyond the single request and catalog every validation rule and required field in the system. Without a cautious admin, end users are fatigued by system errors and begin getting sloppy. To get past a data-heavy screen, they’ll randomly select values and use the old standbys of “-” or “n/a” to get around thoughtfully supplying information.
If things are getting bad and your exec doesn’t heed your warnings, use a stopwatch to time people creating and editing a record. And if all else fails, make your executive sit and watch someone try to do these actions only to be thwarted by error after error.
Do you know the “shiny object” syndrome I mentioned earlier? The one where leadership runs after the next “silver bullet” SaaS solution?
If people are allowed to buy tools at their own discretion, it will lead to multiple tools that do the same thing. And somehow, they will come out of your budget. This will limit the dollars you have to spend on analytics tools that will help the rest of the business make better decisions.
Matthew Volm, CEO and Co-Founder of the RevOps Co-op, said:
“One big ‘don't’ that I learned is Do NOT say yes to every request for every new tool that someone wants. At one point, I ended up having 3-4 different vendors simultaneously providing us with prospecting data."
But it doesn't stop there. He went on to say:
“This caused big data headaches. We had multiple fields that contained the same data from different systems. Then naturally the question is ‘which one is right?’ Then you can expect to ask ‘are any automation rules in SFDC tied to any of these fields? what happens when I sunset one of these?
“To solve for this, I simply started requiring a business case from folks. It could be one paragraph, but I needed people to tell me what problem we were solving, what the ROI would be, and how they would measure success.”
Your predecessor has done their best to set you up with a solid tech stack, but there are...quirks.
Your marketing automation platform is a bit of a mess, your CRM is usable (but clunky), and customer success is on a different, quasi-integrated system. Now that you’re scaling your sales organization, it’s time to get serious about demand generation and smoothing out hand-offs to customer success.
If this sounds familiar, the company has probably busted the startup bubble and made its way into the growth phase. Congrats! Now the name of the game is scaling efficiently, and by “efficiently,” we mean as low cost as possible.
If you’ve been with your company from the start, it’s easy to remember there was a reason you customized your systems. The customizations may not be useful today, but at one point in time, someone in the organization needed whatever is sitting there now.
As someone new to the organization, it’s really easy to be insensitive. I’ve been frustrated when someone has assumed I was an idiot for building something a certain way. After that experience I began work at a new organization, and I still was arrogant enough to turn around and do the same thing to incumbent administrators.
Over the years, I learned to seek to understand before giving advice. Ask clarifying questions and get to the root of why something was done.
Suppose the original administrator is no longer with the organization. In that case, it doesn't mean an executive doesn't look at the element as their pet project (even if no one is using it). Tread lightly and then set about doing the hard work of proving there is a better way.
You will discover old customizations that aren’t being used. And you will also discover people who insist you keep those customizations no matter how much data you gather to show they aren’t being used.
An anonymous RevOps systems leader shared a real-life example:
“One of the company founders insisted on collecting use cases and workflows people were purchasing the product for--which their product team legitimately needed. The founder was convinced there was only one right way to collect the data. This resulted in three tiers of questions with about 20 options to review under each section.
“Unfortunately, the sales team dealt with the requirement by randomly checking boxes, which made it very difficult to conclusively prove that the data was garbage (if only they’d picked the first box every time!). When the founder started demanding the names of people who said they randomly pick options so he could call and chew them out, I refused and gave up trying to change the process. It wasn’t worth betraying peoples’ trust, and there were plenty of other things to fix.”
RevOps professionals typically have a to-do list many pages long, and no matter how many items they check off, new priorities take their place. It can feel like running in place. Don’t make it worse by fixating on every roadblock. Find another battle worth waging and put your energy there.
Change requests stack up quickly. Before you know it, the simple request you promised an executive you could knock out quickly ends up happening three months later.
Some things are too important to ignore. We’ve seen teams maintain sales compensation using Salesforce exports and Excel files for years, spend days every month manually merging data from disparate marketing systems, and try to make do with CRM reports with limited object joins rather than purchase a more robust reporting solution.
Lance Thompson, Technical Sales Operations Manager at SeekOut said:
“I was using a big complicated series of formulas to manage territories. That’s faster the first time you do it, which seemed like a trade-off for assessing territory management systems and taking two weeks to implement something new. What I didn’t account for was that as the number of territories we were supporting grew, it took longer to maintain. I spent 20 hours implementing my own territory solution using a custom object, process builder, and flows in Salesforce. Every time someone needed a territory change (which could happen multiple times per quarter), I saved five hours with the new solution--which means that within four territory changes, I had made up for time spent on the implementation.”
It can be hard to convince people that a RevOps process improvement is worth putting off, say, refreshing a sales process. The way to convince leadership is by calculating your opportunity cost if you continue to put off addressing technical debt. Using the example above, Lance gained the bandwidth to complete one major project per quarter with the time savings gained with the new solution.
Projects to prepare your business systems for scaling will come at your team fast, and it’s impossible to stay caught up unless your business scales your department to keep up with the organization’s demands. Sometimes, the delay in getting something a business leader perceive as simple to do will inspire them to get “creative.”
This usually means either buying a tool they think will solve their problem or hiring someone to install something for them.
We’re not saying outsourcing is always a bad thing. If you do your homework and hire someone very good, it’s a great way to augment the technical side of your operations team in the short run. However, you can’t expect a fantastic consultant to come in cold turkey and know potential landmines they’ll face during their project.
Dhairya Mistry, Salesforce Technical Lead at Qumulo, put it best:
“I see people succumb to a common mistake: ‘Because my manager/leadership wants this, we need to get this project done now. If system administrators are not available, let’s hire consultants and get it done. No need to involve any of the system admins. We'll figure out everything or the consultants will figure it out for us.
“Before starting a project and hiring an external consultant or consulting firm, make sure you have the following ironed out - Why are you doing the project? Why is this important? What are the business objectives, completion timeframe, time you can commit to helping, and what does success look like? What are the clear and measurable outcomes for the project.
"Last but not least, make sure to involve your System Administrator/Architect so s/he can provide insights about your system setup.”
Contractors are great, but they’re expensive long term. Each time you have to hire a different contractor because your original contractor no longer has time, that’s potentially weeks you now need to spend catching the new contractor up on why your system is the way it is and what has to change. Contractors also have a limited number of hours they can dedicate to your projects.
Hiring a full-time person means 100% of their working hours are dedicated to your company alone. They’re also likely to be at your company a lot longer than a contractor.
Another reason to hire a full-time employee? Overloading your one RevOps professional will lead to missed deadlines, an increase in mistakes as they scramble to do everything themselves, and lost efficiency from forcing a single person to try to master everything. It’s more efficient to hire people with expertise and split the workload. Otherwise, your RevOps professional will burn out and leave.
Matthew Volm, CEO and Co-Founder of Funl, says:
“I spent two quarters trying to get free trial form submissions to flow through Segment into Marketo and SFDC to only be foiled. Every. Single. Time. And then we hired a RevOps manager who comes in and gets it set up in less than two weeks via Zapier.”
Switching to something new can seem daunting, but it's essential to be objective. Not selecting a different vendor may mean missing out on the ability to scale quickly. Emotions can get in the way when you’re asked to give something up you've built from day one.
Sometimes the right answer is to stay the course, but make sure you’re doing it for the right reasons.
An anonymous RevOps systems leader said:
"A few years back, I had an opportunity to assess whether a growth stage company kept their existing marketing automation platform, moved to a new platform, or blew up their existing instance and started over with the same vendor. The system's process flows were a mess, but their integrations were working exactly as they should. Seeing that this wasn't the case with other vendors, I chose to overhaul the processes in the existing instance.
“I regretted my decision pretty quickly. We could have replicated the integrations easily in that particular tool and started with a clean slate. Instead, we saved time on integrations but had to investigate every program in the system to figure out what it did, and we were stuck with over 300 junk fields because the vendor didn’t allow users to delete unused fields.”
Enterprise companies have process maturity and larger budgets on their side. That said, they are prone to clinging onto inefficient processes because “that’s how it’s always been done” and overlooking smaller, more innovative vendors by adopting systems their experienced employees have already used.
Companies that have innovative RevOps technology professionals willing to think outside of the box and constantly challenge the status quo are likely to adapt quickly to market changes.
There are some great marketing attribution, sales forecasting, and customer engagement tools out there. One of the biggest missteps we see in larger organizations is the assumption that a spendy tool will bolt onto existing technology and spit out better information than you were getting on your own.
This is only true if you’ve invested in a data management tool that deduplicates, merges, and automatically associates your data. You’ll also need to integrate all relevant tools, which means translating data that uses different identifiers into a universal language. This takes a lot of time and skill, but it’s the only way you’ll get insights rather than misleading metrics.
An expensive marketing analytics tool without paid advertising spend data, web interaction tracking, and a way to align person data with accounts (particularly if you’re a B2B organization) is wasted money.
The other major issue that will prevent you from getting meaningful insights is a lack of CRM user adoption. If your system is difficult to use (too many validation rules, field sprawl, clunky process flow, etc.), your managers don’t enforce usage, and your compensation plan doesn’t have an element requiring CRM usage, your CRM adoption will be low.
An expensive forecasting tool leveraging machine learning is useless if your sales team refuses to use your CRM.
You can do a lot with a CRM, and it will give you a lot of great information (provided adoption is good). However, it doesn’t make sense to integrate every field of every system with your CRM. Your sales team won’t care how much digital ad spend goes into a given campaign. Your customer success team won’t care whether a promotional item was delivered on time.
Smaller organizations may force all users to be in one system and keep everything in a single location. Even they won’t typically integrate LinkedIn data or Google analytics information.
As Jack Robbins, Data Analyst at Notion, said:
“One of the biggest mistakes is trying to turn SFDC or your CRM into data warehouses. I've experienced it, and it is not fun! This is not news to many folks, but SFDC isn't an analytics tool. Yet many organizations attempt to make it into one and create a ton of tech debt and clutter. At the enterprise level, I think it makes more sense to take the sophisticated approach and combine everything at a data warehouse layer (we use Fivetran to export data into Snowflake then visualize the data with Looker). It’s more flexible, more robust, and tool-agnostic; plus you can always leverage a tool like Workato to push data back into SFDC or MA tools, or leveraged embedded analytics in SFDC as well (i.e., embedding a Looker dashboard on a page layout).”
Leveraging a data lake or data warehouse and having a data transformation layer can make things much easier on your system admins, data analysts, and end-users. Instead of gating your systems from creating duplicates, put processes in place to merge data passively. Instead of making people manually enter dates, automate as much as you can.
The easier you make your CRM to use, the better your data will be in the end.
As we mentioned above, inconsistent data definitions can cause major issues. This was my major motivation for embracing a move to Revenue Operations after years in sales, marketing, and customer success organizations.
Watching marketing and sales duke it out over who’s at fault for a bad quarter using the same metrics with different data drove me up a wall.
Creating a standardized definition and centralized reporting system frees analysts from the burden of proving their superiors right and frees them up to focus on adding strategic value. They can spot patterns and dig into why something may be happening, and even come to the table with potential solutions.
I’ve been in organizations that chose to centralize reporting in IT. There are benefits. I didn’t have to write a hundred lines of SQL code to normalize titles. There were also drawbacks. Every time we added a new system, we had to wait for weeks to integrate. A change in the CRM or marketing automation tool took diligent coordination between multiple teams.
The wait time was doable because I still had autonomy. We had socialized data definitions, and as long as we plugged into the same source, we could use a BI tool to summarize the information as we chose.
Business Intelligence teams are amazing, but it’s not possible for everyone to be an expert in everything. The BI team should either allocate a resource dedicated to your department to learn the business processes and context or allow an analyst on your team to dig into the data. Preferably the latter due to ad hoc needs with a quick turnaround time.
As Jack Robbins, Data Analyst at Notion, said:
“Automatic task logging is very important if you want any sort of reliable analytics around sales productivity, lead/inbound SLA, or even funnel tracking. Just as important is actually enforcing the reps to work within those systems. Having Outreach is great, but if you still allow your reps to schedule meetings on Google calendar or other tools external to Outreach--or make calls without Outreach tracking--then you’re as lost as if you didn’t have Outreach to begin with. Strong process enforcement is critical to actually deriving the value from these tools. Your data, and thus your insights, are only as clean as your process hygiene.”
Usability, management support, and compensation are the three pillars of system adoption. Without adoption, your data has too many gaps to be useful. If pushing them into using a single system doesn’t work, leadership should either decide to find a tool that will be adopted or allow multiple points of integration (for example, Outreach integrated with Salesforce and an email plugin to catch interactions not logged in Outreach).
During launch meetings, our most intense debates were about making our internal processes easier versus streamlining the customer's experience.
It’s tempting to require a great deal of information to progress an order or to push implementation earlier in the sales cycle. Is it better for the person doing the information to gather the data? Will something be lost in translation? Will the implementation specialist just have to go back and ask more questions?
Always weigh customer satisfaction more heavily than internal efficiency. You may still opt to streamline your process, but the tradeoff has to be worth it.
A lot goes into making a system scale with the business. It can take a great deal of personal innovation and investment. Which can make it difficult to see when it’s time to throw out your legacy system and make an upgrade.
The pain of switching systems is pretty steep, but if adoption is dropping off and you’re hearing of other systems making traction in the market, it’s worth investigating your options.
When it comes to marketing automation platforms, there are a few that function pretty similarly. If the rates are climbing and your existing system is a bit of a mess (let’s say the existing platform doesn’t allow you to delete custom fields and you’ve had it several years), it may be worth making a change.
Being part of a large company makes it easier to see how large organizations get complacent. If they have a large deal of market share and what they’re doing is profitable, it’s easier to stay the course than make changes that may or may not go over well with the existing customer base.
The same can be said of large technology vendors. While some are known for continuously innovating, it’s not the general rule. Smaller vendors can be scrappy and are more willing to work with companies (let’s be honest, especially large companies) to meet their unique requirements.
