2025 was the year HubSpot stopped being “a tool I implement” and became something closer to a mirror.
It reflects the organisation back at itself, clearly and without much mercy. If internal communication is messy, the data becomes messy. If teams don’t trust each other, workflows grow defensive and over-engineered. You can’t hide operational dysfunction inside a CRM. The system just exposes it.
Across founders, scale-ups, agencies, RevOps teams, Salesforce-heavy orgs, and teams still figuring out who owns what, the same patterns kept showing up. Different industries, different maturity levels, same underlying issues.
What follows are not best practices in the abstract. These are the lessons that only appear once things are live, messy, political, and carrying historical baggage.
If some of this feels uncomfortable, that’s probably a good sign.
Most HubSpot problems are not HubSpot problems
Very few of the issues I worked on this year were caused by missing features or platform limitations.
They were caused by:
-
unclear ownership
-
conflicting definitions between teams
-
processes that evolved without ever being revisited
-
a reluctance to delete or undo past decisions
HubSpot gets blamed when workflows become unmanageable, but that usually traces back to nobody being willing to say “this logic is no longer valid”.
Properties get created “just in case”, pipelines multiply to handle edge cases, and automation piles up without a shared mental model behind it. The result is not a broken system. It’s a system doing exactly what it was told, just without anyone agreeing on what they actually wanted.
Fixing tooling before fixing thinking just creates faster confusion.
Fewer objects, fewer pipelines, fewer workflows usually wins
One of the strongest patterns from 2025 was how often simpler setups outperformed more elaborate ones.
I consistently saw:
-
single, well-designed pipelines outperform three or four “special case” pipelines
-
task-based confirmation working better than exploding stage logic
-
custom objects adding clarity only when ownership and lifecycle were genuinely clear
One team came to me with four deal pipelines, each with eight or more stages created to handle edge cases. We collapsed them into a single pipeline with five stages and task-based checkpoints. Conversion rates barely moved. Stress levels dropped immediately.
Complexity often feels like sophistication. In practice, it usually just spreads accountability thinner.
If a system only works when one specific person understands it, it is already fragile.
The best HubSpot setups I worked on this year were not clever. They were boring, obvious, and easy to explain to a new hire.
Lifecycle stages are not a reporting afterthought
Lifecycle stages are one of the most abused and underestimated parts of HubSpot.
They are not just labels for dashboards. They quietly drive:
-
automation eligibility
-
permissions and handoffs
-
reporting trust
-
how teams interpret success and failure
Once lifecycle stages lose credibility, everything built on top of them starts to wobble.
Common issues I kept seeing included MQL meaning different things to Marketing and Sales, lifecycle stages being backdated “to make reports look right”, and multiple systems acting as competing sources of truth.
By the time someone says “our reporting feels off”, the real problem usually happened months earlier.
Fixing lifecycle logic is painful. Avoiding it just compounds the damage.
Salesforce sync is a strategic decision, not a technical one
This was probably the most emotionally charged area of work in 2025.
Bi-directional HubSpot ↔ Salesforce sync often sounds fair and balanced. In reality, it introduces constant edge cases, ownership ambiguity, attribution arguments, and quiet reporting drift that nobody notices until it’s too late.
The problem is rarely technical difficulty. It’s stress.
Keeping two CRMs in agreement requires clear ownership, fast decision-making, and ongoing maintenance. When those things are missing, even a “working” sync becomes a source of tension.
If no one owns Salesforce full-time, it should not be treated as co-equal with HubSpot.
That decision alone removes an enormous amount of friction.
Automation should reduce thinking, not replace it
The most effective automation I worked on this year did not try to make decisions for people.
It removed admin, surfaced context, and nudged behaviour without pretending the system was smarter than the humans using it.
Good examples included:
-
Slack alerts that audit behaviour rather than enforce it
-
workflows that prepare context for a handoff, not complete it
-
automation that pauses and asks for confirmation at key moments
Fully automated journeys look impressive in diagrams. In real teams, they often hide broken process and delay learning.
If automation removes accountability, it usually creates more work later.
Training is not optional and it is never one-and-done
Teams do not misuse HubSpot because they are careless. They misuse it because no one explained the consequences of their actions.
In 2025, the most successful teams invested in:
-
phased training rather than big-bang onboarding
-
role-based training rather than generic feature tours
-
revisiting fundamentals once real usage patterns emerged
Training is not about showing buttons. It is about building shared mental models.
Good training massively reduces rebuild work later. Poor or rushed training guarantees it.
The biggest learning was saying no earlier
This was the hardest and most valuable lesson of the year.
I got better at:
-
saying no to ownership models that created constant stress
-
stepping away from setups where responsibility was structurally unclear
-
being honest about where I do my best work and where I don’t
Not every technically possible solution is a good one. Not every engagement is worth the long-term cost.
Clear boundaries led to better outcomes for clients and better work for me.
Final thought
2025 reinforced something I now strongly believe.
HubSpot works best when it is treated as a system, not a toolbox. The more intentional teams are about ownership, simplicity, and truth in their data, the less clever their setup needs to be.
Most of the hard work happens before the first workflow is built. The rest is just execution.
Services
Website design
HubSpot consultancy
Podcast production
Salesforce consultancy
Social media management
Marketing strategy
Contact
To get in touch, please drop me a line at tom <at> tjmdigital <dot> com