Modern go-to-market teams didn’t just show up at UNBOUND for AI hype. Over and over, conversations circled back to the same set of operational headaches: systems that don’t talk, data nobody trusts, and sales teams quietly opting out of HubSpot.
The pain is universal, but it’s also fixable.
HubSpot integration challenges are the recurring operational problems that stop CRM, ERP, and finance systems from sharing clean, reliable data. They show up as disconnected tools, reports nobody trusts, and sales teams avoiding the platform, quietly draining revenue and blocking any serious AI or automation gains.
On the stand, this came up more than anything else. Nearly every conversation involved an ERP or finance system – NetSuite, Sage Intacct, Snowflake, niche tools like WoodPro or Quantum – sitting completely outside HubSpot. In a few cases there was a connector, but it was brittle, third-party, and effectively abandoned. The result was the same: order and revenue data lived elsewhere, and HubSpot was reduced to a glorified marketing database.
This isn’t just a UNBOUND problem. Research from Brickclay shows the average enterprise runs around 900 applications, and only 29% are connected. MuleSoft’s Connectivity Benchmark, summarised by Pivot BI Analytics, has integration stuck in the high‑20% range for years. When three‑quarters of your tools don’t talk, misalignment isn’t a surprise, it becomes the default.
The single most painful pattern we heard was simple: “We don’t trust the data.” Leaders told us they had dashboards in HubSpot, finance reporting in an ERP, and offline spreadsheets trying to reconcile the gap. Everyone knew the numbers disagreed, and no one could say which system was right.
Some companies have effectively given up on syncing NetSuite into HubSpot at all. Others are manually reconciling order data, or running separate in‑store and online records that never match. When that’s the backdrop, every report becomes a negotiation and every forecast a guess.
Externally, the picture is the same. According to Brickclay, 80% of organisations name data silos as the biggest barrier to their automation and AI goals. That tracks with what we heard at UNBOUND: people didn’t lack AI ideas; they lacked a single, trusted view of customers and revenue they’d be willing to feed into AI.
The relief point is clear and concrete: define what “source of truth” means for each core object (company, contact, deal, subscription), then design integrations to honour that contract. When teams agree which fields in which system win, conflicts can be resolved once, instead of in every weekly pipeline meeting.
Another recurring theme on the stand was a sharp split between teams. Marketing teams were enthusiastic about HubSpot. Sales teams? Often in Salesforce, spreadsheets, or just ignoring the portal entirely. Leaders framed this as a tooling debate, but most stories sounded like change management and trust problems.
One person we spoke to mentioned that their 300‑person sales teams barely log into HubSpot, and another had a portal that had “decayed over 10 years” to the point where new reps avoid them. That aligns with external data: Salesforce’s State of Sales report, referenced in a sales orchestration piece by Heyreach, finds reps spend only 28% of their time actually selling. The rest is lost in coordination, admin, and working around systems that don’t fit the way they sell.
At UNBOUND, the most honest comments were variations of: “If my reps don’t see their pipeline and commissions reflected accurately in HubSpot, they just won’t use it.” That’s not laziness; it’s rational behaviour. Until the CRM mirrors how revenue is really created – from lead to cash – adoption will always lag behind licences.
A lot of conversations weren’t about ripping Salesforce out, but about making Salesforce and HubSpot live together without constant firefighting. Marketing was happy in HubSpot, another team lived in Salesforce, and nobody wanted to blow up what was already working for them.
We heard this repeatedly: “Salesforce is our source of truth, but we want HubSpot to power the front end.” Others had admissions teams in Salesforce and marketing in HubSpot. The real question wasn’t “which CRM wins?” but “how do we design the data model so everyone can work where they’re strongest without breaking reporting?”
The healthiest examples shared a common pattern. They explicitly documented which fields master in Salesforce, which in HubSpot, and how records move between them. Instead of trying to synchronise everything, they focused on a small set of shared identifiers – the company, the job, the subscription – and made sure those matched every time. That mirrors the “four fields must match” approach described by Pivot BI Analytics.
Another pattern we heard in the UNBOUND corridors: reporting that takes days instead of clicks. Teams described spending hours pulling CSVs from multiple systems, cleaning them in Excel, and manually assembling board reports. One company we spoke to said it takes two days to build a single sales report.
That time cost isn’t just annoying; it’s a direct drag on revenue. In the Heyreach article citing Salesforce’s data, companies with poor sales and marketing alignment see 10% or more revenue loss every year. Slow, manual reporting is usually a symptom of that misalignment – Centralise stitching together what the systems should provide automatically.
The fix isn’t “build more dashboards.” It starts with defining the questions the business actually needs answered (“Which channels create opportunities that close?”, “Where does margin leak between quote and invoice?”) and then aligning integrations and properties so those questions can be answered in one place, reliably, every time.
None of these challenges are new. We've seen them across dozens of portals – disconnected ERPs, duplicated data, sales teams avoiding HubSpot, Salesforce coexistence, brittle third‑party connectors, and reporting nobody fully believes.
MuleSoft’s long‑running benchmark, echoed in Pivot BI Analytics, explains why they persist: even as companies add more apps, the percentage that are actually integrated stays stuck around 27–29%. Integration isn’t a project you “finish”; it’s an ongoing design problem.
Breaking the cycle means treating your revenue system like a product, not a pile of tools. Start small: pick one or two critical journeys (for example, lead → closed‑won → invoiced - or complete our free audit to work out where to concentrate) define the systems and fields that matter, and make those flows boringly reliable.
That’s where clean HubSpot portals, trusted reports, and genuinely useful AI start – not on the expo floor, but in the quiet, deliberate work of making a few key systems talk properly, every time.