I have created a Jotform assessment that is 245 questions. I need MAKE to score each section slightly differently and then pass along the scores/results to Claude API to create a report and eventually connect to GoHighLevel as my CRM. I have attempted multiple scenarios in MAKE, but I cannot get the scoring correct. Need to get this up and running with all connections. Willing to pay for services.
1 Like
Hello @Kerry_Flowers , welcome to make.com community, I have worked and have experience with Make.com and l will love to collaborate with you on this you can schedule a call Here and you can checkout my upwork profile Here, for my pastworks and certifications
Hey Kerry, This is definitely doable in Make.com. I can help you properly structure the scoring logic for each section, process the results, and then pass clean data to Claude API for report generation, followed by integration with GoHighLevel.
The key here is setting up the scoring logic correctly and making sure the data flows cleanly between Jotform, Make, and your CRM.
Quick question: are your section scores already defined in Jotform, or should we handle all scoring logic inside Make?
Check my profile: https://www.upwork.com/freelancers/farhana401
Book a quick call:Calendly - Automaxion
1 Like
Hi Kerry, welcome to the community.
A 245 question assessment with different scoring weights can definitely get tricky inside Make, especially if sections need to be calculated differently before sending the results to an AI model.
One way this is typically handled is by structuring the scoring logic first (either with grouped variables or an aggregator), then passing the final section scores to Claude to generate the report, and finally pushing the results into GoHighLevel.
I’ve worked on similar Make automations involving large form assessments, AI generated reports, and CRM integrations, so I’d be happy to help you get the scoring logic and connections working properly.
You can also check some of my work here in my portfolio website.
Feel free to DM me here if you’d like help troubleshooting the scenario or setting up the full workflow.
Best Regards,
Folafoluwa Stephen
Fola
Folafoluwa - thank you for your interest! I am using asomeone already!
Farhan, thank you for responding. I have reached out to someone already. I will follow up with you if this does not work out. Thank you.
Hi @Kerry_Flowers! I see you found someone already. If you ever need additional help, my team and I would love to help you out.
You can schedule a call in here: Automation Discovery Call | Felipe Saucedo | Cal.com or check our website www.aspirity.com
Hi!
I’ve reviewed your Jotform assessment structure. Scoring 245 questions across multiple sections in Make.com can quickly become a “spaghetti” workflow if you’re using native filters. I’m an automation engineer, and I specialize in building high-logic backends for complex assessments.
How I will fix your scoring and reporting:
-
Centralized Logic: Instead of dozens of modules, I’ll use a single Custom JavaScript (Function) or JSON Array Aggregator to map your 245 questions to their specific sections. This ensures 100% accuracy in the math before the data moves forward.
-
Weighted Scoring: I’ll implement the “slightly different” scoring rules for each section (whether it’s percentage-based, raw scores, or weighted coefficients) directly in the data flow.
-
Claude API Orchestration: I’ll feed these clean, categorized scores into Claude 3.5 Sonnet with a structured system prompt. This will generate a professional, contextual report based on the specific results of those 245 answers.
-
GoHighLevel Integration: Finally, I’ll map these results to Custom Fields in GHL, ensuring the generated PDF or report text is attached to the contact record for your sales team.
Why me?
-
Systems Thinker: I don’t just “zap” things; I build scalable backends. My own website and booking system are fully automated using these same tools: https://mikedevai.netlify.app/
-
Reliability: I build with Error Handling. If a field is missed or the API hangs, the system will retry or alert you instead of losing the lead.
Let’s jump on a call and get your math right:
-
WhatsApp: +375293761570
-
Telegram: @hely_chatbots
Hi Kerry,
Are you still looking for someone to help with this?
I have worked with Jotform, Claude and GoHighLevel in the past so this type of build is well within my wheelhouse.
Drop me a message if you want to discuss scope and pricing
You can reach out to me on my email here
Colin
The scoring layer is the piece I would isolate before wiring the CRM. With 245 questions, Make can get brittle if the section math is spread across routers and filters.
I would normalize the Jotform answers into a section-keyed payload first, run deterministic scoring in one clear step, and send Claude only the scored sections plus the exact report rubric. That keeps Claude from inventing math and makes the GoHighLevel handoff easier to test.
For a first pass I would prove: one completed form, correct section totals, the Claude input/output schema, missing-answer handling, and a logged CRM handoff path. TinyOps Studio can do the fixed $149 workflow audit and return the scenario map plus the failure points to fix. If you want us to build the first working version, email [email protected] with a sample Jotform submission and the scoring rules.
Hi Kerry. I can help get the Jotform scoring logic working in Make before expanding the full report and CRM flow. I would start with one paid slice: normalize the 245 answers, implement scoring for one section, send a controlled prompt to Claude, and return a test output/report. Once that is correct, the same pattern can be extended safely.
Hi Kerry,
This is exactly the kind of multi-logic scoring build I enjoy working on. Jotform + Make.com + Claude API + GoHighLevel is a stack I know well.
The key challenge — getting section scoring right before sending to Claude — is a data structuring problem inside Make. Here is how I would approach it:
Step 1 - Structured scoring in Make: Use an Iterator + Array Aggregator to loop through all 245 question responses and apply per-section scoring rules cleanly. Each section gets its own logic (weighted, raw, percentage) without the scenario becoming unmanageable.
Step 2 - Clean data into Claude API: Pass structured JSON (section names + scores + key responses) into Claude with a consistent system prompt. Claude then generates a professional report based on actual scores.
Step 3 - GoHighLevel sync: Map final scores and the generated report to GHL custom fields and attach to the contact record automatically on form submission.
I have built Make.com + Claude API pipelines producing structured AI reports for clients — including a Google Ads performance pipeline that achieved +797% increase in tracked data using the same architecture.
A few questions to scope this:
- Are the scoring rules already defined per section (e.g. sum, weighted average, percentage)?
- What format do you need the Claude report in — plain text, HTML, or PDF?
- Real-time push to GHL on form submit, or batched?
Happy to fix the scoring logic first as a scoped starting point.
Best,
Son Le | Make.com Automation Specialist
Email: [email protected]