Choosing a HubSpot RevOps consulting partner is a two-year decision. Get it wrong and you're rebuilding workflows, re-cleaning data, and explaining to leadership why the reports still don't match reality.
Most evaluations get this wrong. Buyers compare partners on tier, reviews, portfolio, and price. Then they pick the one that pitched best. Eighteen months later, they're sitting in the same meeting, running the same audit, with a different logo on the invoice (assuming they are still on HubSpot).
The evaluation isn't broken because the criteria are wrong. It's broken because it doesn't ask the questions that actually predict fit.
Why most RevOps evaluations fail
Every HubSpot partner will tell you they specialize in RevOps. Every one will show you a case study, a certification badge, and a proposal that hits your stated needs.
Your stated needs are the problem.
Whatever you wrote in your RFP or your pre-call needs is your self-diagnosis. It's what you think is broken. Most of the time, the thing you think is broken is a symptom of something else, something upstream that you can't see from where you're sitting. A partner who quotes off your self-diagnosis is quoting off the wrong problem.
The right partner runs a real diagnostic first. They check data integrity, workflow health, property architecture, and cross-hub behavior. They separate what's structurally broken from what looks broken but isn't. They quote off what they find, not what you told them.
That's the difference. Everything else is honestly just packaging.
A good diagnostic that is genuinely valuable for your business isn't (and shouldn't be) free. Partners who offer a free audit in exchange for a shot at the business are auditing in a vacuum, often with no signed agreement protecting either side! Your data exposed without contractual guardrails. Also, that's an audit whose findings are quietly biased by the incentive to win the sale. That's irresponsible on multiple levels.
Better model: pay for the diagnostic as its own engagement. Signed agreement, defined scope, fixed price. Then decide whether to move forward on the build. Or cover the majority of the diagnostic in the Kickoff Call and first couple weeks. That's what Fission does.
The four questions that predict fit
You don't need a 40-question RFP. You need four questions. Each one tells you whether a partner is working from evidence or from your assumptions.
1. What do they do before they build anything?
This is the most important question of the four. Ask any partner to walk you through the first two weeks after signing, and listen to the order things happen in.
There are two ways to get this wrong, and buyers usually only recognize one of them.
The first is the partner who never opens the portal. Kickoff, discovery calls, a scope-of-work handoff to the delivery team, then the build starts. Everything they know about your operations came from what you said in a sales call, which means they're building against your self-diagnosis.
The second is the partner who opens the portal and nothing else. You get a findings list. Five hundred custom properties. Workflows still owned by people who left the company. Dashboards built last year that nobody has opened since. All of it accurate. None of it a diagnosis.
Two portals can look identical and mean opposite things. Five hundred custom properties, consistently named and documented, means someone owns naming standards and has the time to enforce them. Five hundred properties, half of them undescribed and inconsistently named, means several people built independently and nobody ever retired anything. Same count but each points to a different problem and a different fix. The portal can't tell you which one you're looking at but the conversation focused on business context can.
The right partner does both. The partner asks about the business first: what you sell, who owns which part of the system, what the team actually does between calls, what broke last quarter that nobody escalated. Then they go into the portal to validate: data integrity, workflow health, property architecture, whether pipeline stages have real entry/exit criteria, how the hubs behave against each other. The audit dimensions are the same either way. What changes is whether the partner has anything to interpret them against.
Two follow-ups separate good answers from rehearsed ones.
Ask what would change the partner's recommendation. A partner running a real diagnostic answers immediately, because they know which of their findings are confirmed and which are still hypotheses waiting on something you haven't told them yet. A partner reading you an inventory has nothing to say, because an inventory contains no hypotheses.
Then ask whether the person doing the diagnose is the person doing the implementation. Most of the market sits on one side of that line. Strategy consultancies diagnose, write the recommendation, and hand implementation back to the same overwhelmed team that couldn't solve it in the first place. Build shops take requirements and execute them competently without ever asking whether the requirements are right. When those are two different companies, or even two different teams inside one company, the diagnosis degrades on the way to the build and nobody owns the gap.
2. How do they sequence the work?
Ask them to describe a project in three phases and tell you what determines the order.
If they can't answer, or if the sequence is driven by what's easiest to build first, walk. A partner without a sequencing methodology will build the flashy thing first because it wins the QBR. Six months in, you find out the flashy thing sits on top of broken data.
A real sequencing answer has named criteria competing against each other, and the partner can tell you which one won. Is anything decaying right now, where waiting means losing it permanently? Is there a regulatory or contractual deadline that overrides everything else? What has to exist before the next thing functions at all? How much change can your team absorb at once? Is there something small and visible to start with that buys the rest of the work internal support?
The first two are the only ones that win. Everything else is a judgment call the partner should be able to defend with their strategic expertise.
Underneath that sits a dependency order that rarely changes: data before automation, automation before reporting, training last. A partner who builds reporting before the data feeding it is governed is building something you'll stop trusting within a quarter, and you'll be the one who has to tell the exec team to stop using the dashboard.
Then, push on the part most partners skip. The right build for a company with a dedicated HubSpot admin is the wrong build for a company where one person owns HubSpot alongside three other jobs. Ask how their recommendation changes based on who maintains it after they're gone. If the sequence doesn't move when that answer changes, you're looking at their standard template with your logo on it.
Main takeaway: every bit of context your partner compiles can, and should, shift their recommendations and build approach.
3. How do they treat AI enablement?
Ask directly: "What's your approach to AI in HubSpot?"
If the answer is a list of tools they'll set up (Breeze this, agent that, workflow the other), they're selling the pitch instead of assessing the readiness. AI doesn't fix broken operations. It amplifies them (shoutout to our homepage).
The right answer starts with your data and your daily processes. It sounds like: "We assess whether your CRM is trustworthy enough to support AI before we scale AI into it." Then they walk you through what they check, and some of it won't be technical. Whether property values are accurate is a portal question. Whether your sales team logs activity in HubSpot instead of working around it is not. That one is a question about your organizational processes, and it determines whether anything AI produces here is worth reading.
Listen for what they think AI is for, too. A partner selling AI as a way to need fewer people is selling you a headcount story that operations may or may not be able to support. The better and more realistic version: your team already has good judgment, and AI raises the floor so the newest person on the team starts their day closer to what your best person would have done.
A partner who understands this will hold the line on foundation work even when the pressure from your board is to skip it.
4. What happens when they leave?
Ask them what they hand off at the end of an engagement.
A shared Google Drive with some Loom videos is a memory test for whoever inherits the portal. A real handoff includes SOPs for every process they built, training material your team can reference without them, and a governance model that explains why things are configured the way they are.
Then ask a sharper version of the same question: what does the last two weeks of a project look like? In a healthy engagement, the final stretch is training, an SOP, or a walkthrough. It's the point where "we built it" turns into "they use it." A project where every build item got completed and no handoff ever happened didn't actually finish.
Most engagements fail here. Not during delivery, but at handoff. If the partner can't describe what the end looks like, they haven't planned one.
The question they should be asking you
A partner running a real diagnostic will ask you things you weren't expecting, and some of them will feel like they're about you (ie. your business and team) rather than about your portal.
Does your leadership team use HubSpot themselves? Will they look at reports that come out of it? Will they require the sales team to log activity there?
Three no's and the best build in the world gets worked around inside a quarter. We've watched it happen. The configuration was right, the data was clean, the training happened, and the team quietly went back to the spreadsheet because nobody above them ever opened the dashboard. That's the strongest predictor we know of for whether operational work holds, and there's no way to see it in a portal audit.
Cultural and organizational dynamics are just as important for your partner to know and it's concerning if they don't ask.
The other half of this is smaller and damages more engagements than it should: YOUR commitments as the client. There are always things a partner depends on the client for such as approvals on definitions, receiving company documents, or providing critical feedback. These project blockers can result in work quietly going away if no one is overseeing it and holding the whole team accountable. A partner who tracks your side of the work as explicitly as their own, and raises a stalled item by name in the meeting, is protecting the outcome. A partner who lets it go silent is protecting the relationship, which is a worse trade for you because the "thing" you all agreed to work on gets forgotten.
So the evaluation runs in both directions. The partner is also assessing whether you can absorb what they build. If nobody asks you that question, they're planning to do the work alone, and work done alone leaves when the partner does.
The red flags that most buyers miss
1. They price deliverables off your list of requests
Start with the shape of what they quote. A partner who prices a fixed list of deliverables straight off your RFP has locked the scope to your self-diagnosis before anyone has looked at anything. Every correction after that becomes a change order, which means the partner is now financially rewarded for your original guess being right.
The alternative isn't a vaguer quote. It's a quote for a level of engagement, where the price reflects the capacity you're buying and the work inside it gets set once the better, more contextual diagnostic has happened. That's a partner who can afford to find out they were wrong. If someone can't price the engagement without the full deliverable list in hand first, they need your self-diagnosis to hold up, and it usually doesn't.
2. Their case studies count what got built
"We built 40 workflows and 15 dashboards" tells you nothing about whether the client's operations improved. Ask what changed for the business. If they can't answer without going back to the client, they never got close enough to know.
3. They lead with AI before they've seen your data
A partner who leads with AI capability before understanding whether your data and operations can support it is selling a pipe dream. A partner who assesses the data and underlying processes first and then explains where AI fits is consulting. The same tell shows up in sequencing: if there's no articulated framework for deciding what to fix first, the order defaults to whatever's easiest or whatever wins the next QBR.
4. They get quiet about documentation
Some partners get quiet on SOPs and handoff processes because they don't want you to be able to leave. That's the wrong incentive pointed at you for the length of the engagement. The right partner wants you running without them and prices the relationship accordingly.
You want a partner that can keep working on projects that truly help the business. Proper documentation and handoffs create team alignment and the environment for scalability.
5. The agreement can't be re-sequenced
Read the agreement for flexibility, not just length. Operational work compounds, and the outcomes that really change the business can take months. So a longer term isn't always the warning sign that buyers treat it as. What matters is whether priorities can move inside the scope. Nine months in, something you agreed to in month one will have stopped mattering, and you should be able to say that in a meeting and see the sequence change that week. If redirecting the work takes a renegotiation every time, the agreement is protecting the partner's revenue forecast rather than your outcome. The right partner works alongside your team as new priorities come up, and leads the charge on how work gets sequenced when priorities change.
And watch how your partners handle the first stretch of the engagement, which brings us to the part most buyers get backwards.
The "audit trap"
The free audit is an easy flag to spot. A partner offering to go through your portal for free in exchange for a shot at the business is auditing in a vacuum, usually with nothing signed protecting either side, and producing findings shaped by the incentive to win your business. You get portal access handed over for free and an assessment you can't trust.
The paid standalone audit looks like the responsible alternative. It usually isn't.
Plenty of partners will sell you a six-week assessment for $5,000 and deliver a deck at the end of it with a long list of homework for you. The findings are real. They're also, in most cases, things you already suspected, written down more formally. Six weeks in, nothing has been fixed. You've paid for a diagnosis and you still have to negotiate a separate agreement to act on it. And at that point, the partner has every incentive to make the findings list long.
The one exception I'll give to the paid audit approach is if there is mutual agreement from the start that your side will be doing the implementation. But that's rare because the whole point you're talking to a partner is because you don't have the time or expertise to build it right.
The model that actually works treats the "audit" as a foundational step of the entire engagement.
At Fission, our audits are our kickoff calls. The portal review and context gathering happen concurrently, because neither one means much without the other. By the first recurring meeting after kickoff, work is already being delivered, and the scope is written against what we found rather than against your initial wish list of projects.
A prospect who had already sat through pitches from three other HubSpot partners told us this on our first call:
"You're the only one that hasn't recommended jumping into our account yet to see what's there."
And they signed with us because every other partner treated portal access as the first step of the sales process which we fundamentally disagree with. Going into a portal cold is how you end up confidently solving the wrong problem.
So there are three ways a partner can handle the first phase, and only one of them has you paying for real work instead of paying for a fancy deck.
How to actually run the evaluation
Start by being honest about the full set of options, because going with an agency is only one of the options you have to pick from. The real alternatives are hiring someone internally, using a generalist agency you already work with, and, of course, doing nothing. A senior internal ops hire runs $120K to $185K fully loaded with a 3-6 month ramp, and the organizational knowledge leaves when they do. A generalist agency will be inadequate at operations because it isn't their core competency. They may be a full-stack marketing agency that knows HubSpot; doesn't mean they know operational strategy and execution. And finally, doing nothing has a real cost that never appears as a line item. Each of those is the right call in some situation. Know which situation you're in before you start comparing partners.
Then pick three partners, not six. Six is a decision-fatigue trap. Three is enough to see the spread.
Run each of them through the four questions in a 45-minute call. Don't send it as an RFP or a project doc, send it as a conversation. You'll learn more from how they answer than from what they submit in writing, and the answers are hard to fake because each one exposes whether there's a method behind it. Plus, you're removing any "fake it 'til you make it" approach by running it live.
Pay attention to what they ask you. The partner who spends the call diagnosing is showing you exactly what the engagement will feel like. The partner who spends it presenting is showing you they can follow a high-quality script.
Then ask each finalist the same closing questions:
-
What would the first two weeks require?
-
What would you deliver in them?
-
What would I own at the end?
You're looking for a partner whose first two weeks produce both a diagnosis and delivered work, under a signed agreement, at a price you'd have paid for the work alone. If the answer is a six-week assessment with a deck at the end, you're buying a document and a task list for your internal team that's already swamped.
A short note on ranked comparisons
You've probably seen "Best 6 HubSpot RevOps Consultants of 2026" listicles. Most of them are written by one of the firms on the list (and shocker, they're at the top).
We don't publish rankings for that reason. Any comparison authored by a firm that ranks itself in the comparison is compromised. The frameworks in those pieces might be useful. The rankings aren't.
Use the framework above instead :)
The whole point
The right RevOps partner:
-
investigates the business before they investigate the portal
-
sequences the work by what breaks if you delay rather than by what looks impressive
-
treats AI enablement as a downstream capability instead of a headline
-
and hands off SOPs so your team can run without them.
If you want to see what that looks like on your portal, that's what our kickoff call is. No six-week deck, no free audit. The fastest way to find out whether we're the right fit, or whether you should be talking to someone else entirely.
