Seeking Advice on Rebuilding a CRM During Ongoing Business Operations and HubSpot Cleanup Strategies | RevGenius

santiago362
·
·
Led a couple of these. What saved us: freeze net new schema changes for two weeks, lock a single source of truth, and fix naming at the integration layer first, not in the UI. On your specific questions: KPIs and success: we defined it as trust in the number, not field completeness. The main KPI was percent of pipeline where front end and back end reconcile, and we drove that from around 60 to 95 plus. Phases: yes, three. 1) Stabilize: freeze changes, dedupe, stop the bleeding. 2) Re map naming to integration values behind a translation layer so nothing breaks live. 3) Backfill history and turn reporting back on. Timeline: 8 to 10 weeks for a mid size HubSpot, most of it sitting in the re map phase. Cleanup vs ongoing requests: one dedicated owner for the rebuild, all new asks routed through a weekly intake, so foundational work never got preempted by the loudest request. Hope it helps!
👍1
kemal
·
·
I’d treat this as a phased cleanup, not one big rebuild. First define the source of truth for each core field: lifecycle stage, lead source, owner, segment, and integration IDs. Then map front-end labels to back-end values before changing anything. For KPIs, I’d track: % of records with complete required fields duplicate rate unmapped integration values reporting accuracy on key dashboards sales/CS adoption after the cleanup The biggest risk is fixing naming conventions without documenting the new operating rules. Otherwise the same data quality issues come back in 3–6 months.
👍1
Serena S.
·
·
Thank you so much for sharing! Phased approach was definitely where I was leaning but the biggest challenge as you have both highlighted is maintaining data integrity as revops changes over from old to new and GTM teams continue to enter their daily activities.
Sourabh
·
·
We've run a few of these — the hard part is keeping things running while you fix it. One thing that matters early: figure out if it's really a naming/mapping problem or if the object structure itself is broken. If it's mapping, clean in place — faster, lower-risk, integrations stay intact. If the structure's actually broken, a fresh instance can be worth it, but it's a heavier lift so worth confirming first. We've helped a few teams through this kind of HubSpot cleanup at Panoptic — feel free to reach out if you want to compare notes. Sourabh https://panopticanalytics.com/
Ashok A.
·
·
We did this exercise a few months ago. The way we did it was to first create a bible of what's what. Get the entire leadership to agree on it. No more exceptions after that. Changes were done in phases. Leadership was expected to adhere to the changes, Revops flagged any non-adherence in a weekly cadence leadership meeting, corrections were made real-time before moving to the next phase. We did the whole exercise and took us under 8 weeks for the whole revamp. We stuck to the bible, no revisions, just adherence in phases. But creating that master bible took us about 4 long-haul meetings, once it was well worth the effort for the one-time fix. All the best 🙂
Ashok A.
·
·
• How did you define success and set KPIs for the RevOps team? - Success metric was defined based on the quality of data, timeline, adherence and reporting quality. • What timeline did you establish for the cleanup and rebuild effort? - Across 170 sales folks, it was a decent 8 week roll out in phases. L1 thru L4 depending on complexity. Weekly leadership cadence is non-negotiable. Adherence was a new norm. Top down approach. And then roll out was pretty easy to the entire team once leadership agreed upon the bible. • What metrics did you use to track progress (data quality, adoption, reporting accuracy, integration health, etc.)? - Everything essentials for BAU, the non-urgent ones, the ones that had to be recreated or built from ground up were pushed for Phase 2 after the initial 8 weeks. • Did you tackle the work in phases, and if so, how did you prioritize them? - we took the bible seriously. Classifed what was essential, what was optional. That was a great starting point. With Hubspot, leaving the fields already used will certainly help. Just replace with newer, lesser fields on CRM. Never, ever delete fields or data on CRM. This is the first step for the mess to happen. • How did you balance foundational cleanup work against ongoing business requests and stakeholder expectations? - Simple, we did not touch anything that was BAU, we started replacing them with more meaningful fields that are bare essentials. Hope this helps
Serena S.
·
·
Hi Ashok A. great feedback and insights. A few follow up questions for you:
1.
Out of curiosity, what was the structure for your bible and what would you add or take out to improve it in retrospect?
2.
How large was your team for the 8 week turnaround? How was work distributed amongst the team members?
3.
How did you handle issues that surfaced during the rebuild phase?
Ashok A.
·
·
1.
Out of curiosity, what was the structure for your bible and what would you add or take out to improve it in retrospect? -
a.
We started painting the picture of what each one of us needed to have access to, view, edit etc by sales roles.
b.
BDRs, SDR,s AE's, Sales Managers, Directors, C levels, partners, and Board members.
c.
It included everything from views, dashboards, fields that can be edited, forced objective filters instead of subjective answers (Easier to draw metrics/dashboards etc)
d.
This was the core starting point.
Once this was drawn, we wanted to see how far we were with the existing workflows. The gap was too large, we couldn't even achieve 20% from the picture we painted and had to built almost 80% of the fields. The simplest way was to built the flows by reading the data from the existing fields, and not by deleting the fields/data. I strongly recommend that you take a backup on Hubspot before making large corrections. This allows you to restore if things go south. Hope this structure helps.
1.
How large was your team for the 8 week turnaround? How was work distributed amongst the team members? - We were 170 folks in total when this activity happened. The teams ranged between 2 folks to 56 folks across 9 different teams in multiple roles starting from BDR to CEOs.

1.
How did you handle issues that surfaced during the rebuild phase? - There were no major issues as such. We learnt that during the first few days, major issues occur when you delete a field or replace a field. Remember, this will disrupt the entire workflow the teams were used to all these years. I strongly recommend you manually create fields and start replacing like-to-like fields, so both old fields and new fields co-exist.
Once the entire org is happy with the workflow and everything is working as expected and you've given your team to adopt the changes, just delete the old fields and migrate fully after 8 weeks or so.
1.