MELLEKA MARKETING

Bay Street Lending

CRM, AI & Automations Deep Dive
April 2026 through September 2026
96 Total Tasks
12 Categories
6 Months
85 Completed

Executive Summary

Since launching our engagement in April 2026, Melleka Marketing has executed a sweeping infrastructure buildout across Bay Street Lending's entire marketing and CRM technology stack. This deck catalogs every single task completed or initiated across CRM integrations, marketing automations, AI initiatives, conversion tracking, email/SMS nurture sequences, and platform operations.

The work spans the full lifecycle of a modern growth operation: connecting Salesforce as the source of truth, building Zapier integrations to sync leads across platforms, designing multi-step nurture workflows in the Melleka App (GoHighLevel), standing up JustCall automations for missed-call follow-ups, launching conversion tracking across Google, Meta, Bing, Reddit, and LinkedIn, implementing click-fraud protection, and orchestrating a phased rebrand from "Bay Street Lending" to "Bay Street Capital."

Every task listed below represents hands-on execution by the Melleka team, from Salesforce webhook configuration and named credential setup, to HTML email template design, toll-free number provisioning, and ad platform AI automation enablement. This is the full accounting of what has been built, what is live, and what remains in progress.

96
Total Tasks
85
Completed
11
In Progress / Pending
12
Categories
88%
Completion Rate
6
Months Active
Salesforce Lead & Opportunity Results
April 2026 through September 2026

The following data represents the cumulative lead generation and opportunity pipeline results tracked in Salesforce since the start of the engagement. These numbers reflect the combined output of all paid advertising platforms, website lead forms, and AI-driven channels that Melleka Marketing has built, launched, and managed for Bay Street Lending.

Paid Search Leads
383
Total Paid Search Leads
118
Google Ads
91
Bing / Microsoft
1
Facebook
1
Yahoo
172
No Source Attributed
Lead Status Breakdown
Closed - Not Converted (186)
Open - Not Contacted (179)
18
18 Working
186 Closed - Not Converted
179 Open - Not Contacted
Website Opportunities
210
Total Website Opportunities
18
Won Opportunities
192
Open / In Pipeline
8.6% win rate on website-sourced opportunities
ChatGPT / AI-Sourced Leads
8
Total ChatGPT Leads
2
Currently Working
ChatGPT Ads (OpenAI) is an emerging channel that Bay Street Lending was among the first in the lending industry to activate. These 8 leads represent early traction from a net-new advertising platform.
Category Breakdown
Category Completed In Progress Total
Salesforce Integrations & Automations 14 1 15
Salesforce Operations & Data 7 0 7
Zapier & Webhook Integrations 6 1 7
JustCall Integrations 7 0 7
Melleka App / GHL Workflows 8 2 10
CRM & Platform Setup 9 0 9
Email & SMS Nurture Sequences 16 0 16
Tracking, Attribution & Conversion 11 2 13
Ad Platform Operations 4 0 4
AI Initiatives 2 2 4
Rebrand & Identity 5 0 5
Client Communications & Reporting 5 0 5
Total 85 11 96
Salesforce Integrations & Automations
CRM connectivity, webhook configurations, and automated workflows between Salesforce and marketing platforms
15
1
Connected Salesforce to the Melleka App (GoHighLevel) and ensured all leads flow into the Melleka App
Established the foundational CRM-to-marketing-platform integration, enabling Salesforce lead records to automatically sync into the Melleka App for nurture workflows, SMS outreach, and email marketing. This was the critical first step in building the full automation stack.
Done
2
Set up a Salesforce automation to send data to a webhook once a new contact is added to the TrustPilot list
Configured a Salesforce Flow that fires whenever a new contact record is added to the TrustPilot review list. The flow packages the contact data (name, email, funded amount) and sends it to an external webhook endpoint, enabling downstream automation in the Melleka App to trigger personalized TrustPilot review request emails.
Done
3
Created named credentials and external webhook credentials in Salesforce
Set up the secure authentication infrastructure within Salesforce to allow outbound webhook calls. Named Credentials in Salesforce handle secure storage of endpoint URLs and authentication tokens, ensuring the webhook integrations meet enterprise security standards and can communicate with external systems without exposing API keys.
Done
4
Configured the TrustPilot review automation using the agreed filters from Vadym's list
Implemented the full TrustPilot review request automation in Salesforce based on filters agreed upon with the BSL team. The automation identifies eligible contacts based on specific criteria (funded amount, exclusion flags) and routes them into the review request workflow.
Done
5
Configured review email triggers for contacts whose funded amount is greater than zero and whose TrustPilot exclusion field is false
Built the specific trigger logic that ensures only qualified contacts receive TrustPilot review requests. The automation checks two conditions: (1) the contact's funded amount must be greater than $0, confirming they are a real funded client, and (2) the TrustPilot exclusion boolean field must be false, ensuring previously excluded contacts are not contacted. This prevents review requests from going to prospects, denied applicants, or contacts who have opted out.
Done
6
Implemented Salesforce "Do Not Call" sync when contact is set to DND on the Melleka App
Investigated and built a bidirectional compliance workflow: when a contact is flagged as DND (Do Not Contact) in the Melleka App, the system automatically updates the corresponding Salesforce contact record to "Do Not Call" status. This ensures regulatory compliance (TCPA) is maintained across both platforms and prevents agents from calling contacts who have opted out of communications.
Done
7
Added a custom field for lead status and mapped from Salesforce to create a trigger for email workflows
Created a custom lead status field in the Melleka App that mirrors Salesforce lead statuses, then built the field mapping between both systems. This mapping enables automated email workflow triggers based on status changes (e.g., when a lead moves to "Working" or "Closed - Converted" in Salesforce, the corresponding Melleka App workflow adjusts its nurture sequence accordingly).
Done
8
Turned off PayPal Sent / Check Sent Zap per Maria's request
Disabled the Zapier automation that was sending notifications for PayPal payments and check disbursements, per direction from Maria on the BSL team. This was part of ongoing automation hygiene to ensure only active, approved workflows remain running.
Done
9
Mapped Facebook lead form fields for all active lead forms into their corresponding Salesforce fields
Audited every active Facebook/Meta lead form and mapped each field (name, email, phone, business type, loan amount, etc.) to its corresponding Salesforce field. This ensures lead data captured from Meta ad campaigns flows cleanly into Salesforce without data loss or misalignment, enabling accurate attribution and immediate sales follow-up.
Done
10
Requested and set up Salesforce Marketing Module access for Anthony Melleka
Coordinated with BSL to provision access to the Salesforce Marketing Module, enabling the Melleka team to directly configure marketing automations, email templates, and campaign attribution within Salesforce rather than relying on intermediary requests.
Done
11
Reviewed general Salesforce workflow improvements with the team
Conducted a comprehensive review session examining the existing Salesforce workflow architecture, identifying inefficiencies, redundancies, and opportunities for optimization. Documented improvement recommendations for the broader automation strategy.
Done
12
Went through all 54 existing Salesforce automations to identify email templates that need redesign, along with their triggers
Performed a complete audit of all 54 active Salesforce automations. For each automation, documented the trigger conditions, the email template(s) used, and whether the template needed to be redesigned for branding consistency, mobile optimization, or performance improvement. This audit served as the foundation for the email template redesign initiative.
Done
13
Investigated Salesforce automations and routing logic for referral emails
Audited the existing Salesforce automation rules governing referral email sends. Examined the routing logic to understand how referral sources are tracked, how referral emails are triggered, and whether the current automation accurately reflects the intended business process.
Done
14
Audited the Salesforce automations and routing logic for all leads
Conducted a thorough audit of how leads are routed within Salesforce, including assignment rules, lead status transitions, notification triggers, and handoff logic between marketing-qualified and sales-qualified stages. Identified gaps in the routing logic and recommended improvements.
Done
15
Confirmed whether multiple address-verification statuses exist, and scoped a CRM automation that sends texts, emails, and creates tasks based on those statuses
Investigated the address verification workflow within the CRM to determine if multiple verification statuses (verified, pending, failed, etc.) exist. Based on findings, scoped an automation that would trigger different outreach sequences (SMS, email) and internal task creation based on each verification status, ensuring borrowers receive appropriate follow-up at each stage.
In Progress
Salesforce Operations & Data
Data management, attribution logic, reporting, and operational tasks within Salesforce
7
1
Built improved attribution logic in Salesforce that separates organic leads from paid leads, and requested a review call to walk through it
Developed a new attribution model within Salesforce that accurately distinguishes between organic leads (website form submissions, direct traffic, referrals) and paid leads (Google Ads, Meta Ads, LinkedIn). The improved logic ensures marketing spend attribution is accurate and prevents paid leads from being miscategorized as organic, which was inflating organic lead counts and understating paid campaign ROI.
Done
2
Pulled the channel field in Salesforce into the main column and resolved the attribution issue
Reorganized the Salesforce lead record layout to surface the channel/source field into the primary view columns. This made lead source attribution immediately visible to the sales team without needing to drill into each record, and resolved a persistent attribution tracking issue where channel data was being lost or overwritten.
Done
3
Worked on better attribution tracking in Salesforce
Continued iterating on the Salesforce attribution architecture to improve accuracy across all lead sources. This involved refining UTM parameter capture, strengthening the mapping between ad platform click IDs and Salesforce records, and closing gaps in the attribution chain.
Done
4
Checked on leads and Salesforce data integrity
Performed a data integrity audit of lead records in Salesforce, verifying that lead data was flowing correctly from all sources (Meta, Google, website forms, JustCall), statuses were accurately reflected, and no records were stuck, duplicated, or missing critical fields.
Done
5
Checked Meta leads being converted into website source
Investigated and flagged an issue where leads originating from Meta (Facebook/Instagram) ad campaigns were being incorrectly attributed as "Website" source in Salesforce. Traced the root cause to how the Zapier integration was handling the source field mapping, and implemented a fix to preserve the original "Meta" source attribution.
Done
6
Retrieved operational numbers from Salesforce
Pulled key operational metrics and lead volume data from Salesforce for internal analysis and client reporting. This data was used to benchmark lead flow, conversion rates, and pipeline performance across different lead sources and time periods.
Done
7
Shared the "Funded Contacts" list from Salesforce Lightning and flagged an outstanding data-quality issue with "unknown contacts"
Exported and shared the list of funded contacts from Salesforce Lightning for use in the TrustPilot review automation. During this process, identified and flagged a data quality concern: a subset of contacts were categorized as "unknown" with incomplete or missing data fields, which needed to be resolved before including them in any automated outreach.
Done
Zapier & Webhook Integrations
Automated data pipelines connecting Salesforce, Meta, web forms, and the Melleka App
7
1
Created Zap: Web form submissions to Melleka App (sent instructions to Vadym)
Built a Zapier automation that captures every new web form submission from the Bay Street Lending website and creates a corresponding contact record in the Melleka App. This ensures every website lead enters the nurture workflow automatically without manual data entry. Sent configuration instructions to Vadym for the website-side form integration.
Done
2
Created Zap: Meta lead form submission to email notification
Built a Zapier automation that sends an instant email notification to the BSL sales team whenever a new lead is submitted through a Meta (Facebook/Instagram) lead form. This ensures zero delay between lead submission and sales team awareness, enabling rapid follow-up.
Done
3
Zapped the Meta lead notification
Configured the Meta lead notification Zap to ensure proper formatting, delivery reliability, and correct recipient routing. Verified the notification fires consistently for all Meta lead form variants across active campaigns.
Done
4
Created Zap to transfer closed leads from Salesforce to Melleka App, tag them, and notify Emely
Built a Zapier workflow that monitors Salesforce for leads that move to a "Closed" status. When detected, the Zap automatically creates or updates the contact in the Melleka App, applies the appropriate tag (e.g., "Closed - Converted" or "Closed - Lost"), and sends a notification to Emely on the BSL team confirming the transfer. This ensures the Melleka App always has current lead status data for downstream automations.
Done
5
Created Zap to update Melleka App when Salesforce leads update
Built a real-time sync Zap that monitors Salesforce lead record updates and pushes those changes to the corresponding contact in the Melleka App. This ensures lead status, contact information, and custom field values remain synchronized across both platforms, preventing nurture workflows from sending outdated or inappropriate messages.
Done
6
Evaluated options (including Zapier and Cloud Code) to push attribution data from the website into Salesforce
Researched and evaluated multiple technical approaches for capturing website visitor attribution data (UTM parameters, referral source, landing page, click IDs) and pushing it into Salesforce lead records. Compared Zapier, Salesforce Cloud Code (Apex), and JavaScript-based solutions. Reported findings and recommended approach to the team.
Done
7
Configure the Zapier step to map contacts in the Melleka App once the webhook data comes in from Salesforce
The final integration step for the Salesforce-to-Melleka App TrustPilot pipeline: configuring the Zapier step that receives the inbound webhook data from Salesforce (triggered when a contact is added to the TrustPilot list) and maps it to create or update the corresponding contact record in the Melleka App with proper tags and fields for the review request workflow.
Pending
JustCall Integrations
Phone system automations for missed-call follow-ups, SMS triggers, and cross-platform workflows
7
1
Created a workflow in JustCall to post information to a webhook when a contact doesn't pick up
Built an automated workflow within JustCall that detects when an outbound call is not answered by the contact. Upon detecting a missed/unanswered call, the workflow packages the call data (contact name, phone number, call time, agent ID) and sends it to an external webhook endpoint. This webhook serves as the trigger for the downstream Melleka App automation that initiates SMS and email follow-up sequences.
Done
2
Changed the workflow to send SMS after a not-connected phone call from JustCall to the Melleka App
Modified the existing JustCall missed-call workflow to route the trigger through the Melleka App's SMS engine rather than JustCall's native SMS. This change ensures all outbound SMS messages are sent from the BSL marketing phone number registered in the Melleka App, maintaining brand consistency and ensuring messages are tracked within the centralized communication platform.
Done
3
Created a workflow in the Melleka App to receive data from JustCall when a contact doesn't pick up, create a contact, add a tag, and send an email
Built the receiving end of the JustCall missed-call automation in the Melleka App. When the webhook fires from JustCall, this workflow: (1) Creates a new contact record if one doesn't exist, (2) Applies a "Missed Call" tag, (3) Triggers an automated follow-up email letting the contact know they were called and providing alternative ways to get in touch. This closes the loop on every unanswered outbound call.
Done
4
Wrote text/SMS options and sent to Carmine for approval for the call-failed automation
Drafted multiple SMS copy variants for the missed-call follow-up automation. Each variant was crafted to be conversational, non-pushy, and compliant with SMS marketing regulations. Submitted all options to Carmine for review and approval before activating the automated sends.
Done
5
Created an email for when a contact does not pick up the phone
Designed and built an HTML email template for the missed-call follow-up sequence. The email acknowledges the attempted call, provides the caller's contact information, and offers easy ways for the contact to schedule a callback or submit their information online. The template was optimized for both desktop and mobile rendering.
Done
6
Responded to Carmine about phone numbers in JustCall and the Melleka App
Addressed Carmine's questions about the phone number architecture across JustCall (sales dialing) and the Melleka App (marketing SMS). Clarified which numbers are used for which purpose, how caller ID is displayed, and the compliance implications of using different numbers for outbound calls vs. automated SMS messages.
Done
7
Investigated setting up a workflow for JustCall when contacts don't pick up, to send an automatic text message follow-up
Initial research and scoping phase for the JustCall missed-call automation. Evaluated JustCall's API capabilities, webhook options, and workflow builder to determine the optimal architecture for detecting unanswered calls and triggering automated SMS follow-ups. This research informed the final implementation.
Done
Melleka App / GoHighLevel Workflows
Marketing automation workflows, lead nurture sequences, DND handling, and workflow trigger management
10
1
Created lead nurture workflow: lead form/new contact > 2 min wait > SMS #1 > 3 days wait > Email #1 > 5 days wait > SMS #2 > 2 days wait > Email #2
Designed and implemented the core lead nurture automation sequence in the Melleka App. This multi-step workflow activates when a new lead enters the system and executes a carefully timed series of SMS and email touchpoints over a 12-day period. Each touchpoint escalates in urgency and information density, moving from a friendly introduction to detailed value propositions. The timing was calibrated to maintain engagement without overwhelming the prospect.
Done
2
Corrected workflow triggers for the lukewarm lead sequence: removed the previous trigger and installed a new trigger with a new delay to account for wait time and prevent immediate messaging
Identified and fixed a timing issue in the lukewarm lead workflow where contacts were receiving messages too quickly after entering the sequence. Removed the original trigger, built a new one with an appropriate delay buffer, and tested to ensure contacts receive their first touchpoint at the right interval after being tagged as "lukewarm."
Done
3
Updated Lukewarm Lead Nurture Workflow to include Emails #3 and #4 along with conditions and wait times to exclude leads with status "Working" or "Closed - Converted"
Extended the lukewarm lead nurture workflow from a 2-email sequence to a 4-email sequence. Added conditional branching logic that checks the lead's current Salesforce status before each touchpoint: if the lead has moved to "Working" (being actively pursued by sales) or "Closed - Converted" (deal done), the workflow automatically removes them from the nurture sequence. This prevents marketing from messaging leads that are already being handled by the sales team.
Done
4
Created workflow adding contact as DND and alerting Hunter with an internal notification that someone replied "stop"
Built a compliance automation that monitors inbound messages for opt-out keywords ("stop", "unsubscribe", etc.). When detected, the workflow immediately: (1) Sets the contact's status to DND (Do Not Disturb) in the Melleka App, (2) Removes them from all active automation sequences, (3) Sends an internal notification to Hunter on the BSL team alerting them of the opt-out. This ensures TCPA compliance is maintained in real-time. Sent to David for QA verification.
Done
5
Filtered out people who have been dialed more than once to prevent sending them duplicate messages
Added a deduplication filter to the outbound messaging workflows. The filter checks the contact's call history and suppresses automated SMS/email sends for contacts who have already been dialed multiple times, preventing them from receiving redundant automated follow-ups that could feel spammy or annoying.
Done
6
Disabled text messaging in the outreach workflow
Per direction from the team, turned off the SMS component of the outreach workflow to pause automated text message sends. This was done as part of a workflow refinement cycle to evaluate the performance of email-only nurture before re-enabling SMS touchpoints.
Done
7
Rephrased the outbound SMS message to make it more enticing
Rewrote the automated outbound SMS copy to improve engagement rates. The revised message was crafted to be more conversational, include a clearer value proposition, and encourage a direct response from the recipient. The new copy was tested and approved before deployment.
Done
8
DND the contact once determined that their business is too small or any other disqualifying reason (Dewayne Cowen case)
Implemented a manual DND process for contacts who are identified as unqualified during sales outreach (business too small, wrong industry, not a fit for BSL's lending products). Used the Dewayne Cowen case as the template to establish the standard operating procedure for disqualification tagging and DND application in the Melleka App.
Done
9
Create re-engagement workflow for leads who did not convert
Designing a re-engagement automation sequence targeting leads who entered the pipeline but did not convert. The workflow will re-activate dormant leads with fresh messaging, updated offers, and value-driven content at strategic intervals after the initial nurture sequence has concluded.
Held Up
10
Create Workflow for "Leave a Review on TrustPilot" after identifying the correct contacts
Building the TrustPilot review request workflow in the Melleka App that will receive funded contact data from the Salesforce webhook and send a personalized email requesting a TrustPilot review. Waiting on confirmation of the correct contact list before activation.
Ready
CRM & Platform Setup
Initial platform configuration, sub-account creation, domain setup, phone number provisioning, and system onboarding
9
1
Created sub-account on the Melleka App (GoHighLevel)
Provisioned a dedicated sub-account (location) within the Melleka App for Bay Street Lending. This includes configuring the account settings, timezone, business information, notification preferences, and default pipeline stages. The sub-account serves as the centralized marketing automation hub for all BSL campaigns.
Done
2
Invited clients to the Melleka App
Sent access invitations to BSL team members, giving them visibility into the Melleka App dashboard where they can view lead activity, conversation threads, appointment bookings, and pipeline status in real-time.
Done
3
Integrated Meta on the Melleka App
Connected the BSL Meta (Facebook/Instagram) ad account to the Melleka App, enabling Meta lead form submissions to flow directly into the CRM for automated nurture workflows, tagging, and pipeline management.
Done
4
Purchased and verified toll-free number for marketing SMS
Acquired a dedicated toll-free phone number for BSL's marketing SMS campaigns. Completed the telecom verification process which includes A2P 10DLC registration, carrier approval, and anti-spam compliance checks. This number is now the official sender for all automated SMS communications from the Melleka App.
Done
5
Set up toll-free number infrastructure for marketing SMS, including collecting the EIN and legal business information required for telecom provisioning and registration
Handled the full compliance infrastructure for toll-free SMS: collected BSL's EIN (Employer Identification Number), legal business name, business address, and authorized representative information required by telecom carriers for toll-free number provisioning. Submitted the A2P (Application-to-Person) messaging registration to ensure deliverability and compliance with carrier regulations.
Done
6
Set up email domain for CRM sends
Configured and authenticated the email sending domain for BSL's marketing email campaigns. This includes SPF, DKIM, and DMARC record setup to ensure maximum email deliverability and prevent messages from landing in spam folders. The sending domain is properly isolated from BSL's primary business email domain to protect sender reputation.
Done
7
Discussed email domain strategy for CRM and marketing email sends
Conducted a strategic planning session on email domain architecture. Weighed the pros and cons of porting BSL's existing domain for marketing sends versus purchasing a dedicated sending domain. Flagged that domains used for high-volume marketing email eventually accumulate spam complaints and need to be rotated, recommending a dedicated sending domain to protect the primary business domain's reputation.
Done
8
Obtained backend access from Carmine
Coordinated with Carmine to secure backend access to BSL's website and internal systems, enabling the Melleka team to implement tracking codes, form integrations, and technical configurations directly rather than going through intermediaries.
Done
9
Investigated Salesforce-to-HubSpot migration feasibility
Explored the possibility of transferring CRM operations from Salesforce to HubSpot. Evaluated data migration complexity, feature parity, workflow rebuild requirements, and cost implications. Assessed how much of the existing Salesforce infrastructure could be replicated in HubSpot and reported findings to the team.
Done
Email & SMS Nurture Sequences
Custom email templates, SMS copy, HTML optimization, and nurture content created for automated workflows
16
1
Created Email Campaign #1 for Lead Nurture Workflow
Designed and built the first email in the lead nurture sequence. This email serves as the initial touchpoint after the first SMS, introducing Bay Street Lending's value proposition, highlighting key lending products, and establishing credibility. Written in a warm, professional tone and optimized for both desktop and mobile rendering.
Done
2
Created Email Campaign #2 for Lead Nurture Workflow
Built the second email in the lead nurture sequence. This email deepens the relationship by providing more specific information about loan products, success stories, and clear next steps. Designed to convert warm leads who engaged with Email #1 but haven't yet taken action.
Done
3
Created Informative Email #3 as part of the Lukewarm Lead Workflow
Designed the third email in the lukewarm lead nurture sequence, focused on providing educational, value-driven content rather than a direct sales pitch. The email positions BSL as a knowledgeable industry partner, offering insights about business lending that help nurture leads who are interested but not yet ready to apply.
Done
4
Created Informative Email #4 as part of the Lukewarm Lead Workflow
Built the fourth and final email in the lukewarm lead nurture sequence. This email acts as a soft close, combining educational value with a clear call-to-action. It addresses common objections, highlights the simplicity of BSL's application process, and provides multiple ways to initiate contact.
Done
5
Transferred Lukewarm Lead Nurture Email #1 (Revamped) to Melleka App, optimized for Desktop and Mobile
Took the revamped version of Lukewarm Email #1, converted it to responsive HTML, tested across multiple email clients (Gmail, Outlook, Apple Mail, mobile), and deployed it into the Melleka App's email builder. Ensured proper rendering on all screen sizes, correct image sizing, working links, and consistent branding.
Done
6
Created HTML code optimized for Mobile and transferred Lukewarm Lead Nurture Email #2 to the Melleka App
Hand-coded the HTML for Lukewarm Email #2 with mobile-first responsive design principles. Implemented media queries, fluid image sizing, scalable typography, and touch-friendly CTA buttons. Transferred the completed template into the Melleka App and verified rendering across devices.
Done
7
Redesigned Lukewarm Lead Nurture Email #3 in HTML
Completely redesigned Email #3 from scratch using custom HTML to match the updated brand standards and visual consistency of Emails #1 and #2. The redesign improved visual hierarchy, CTA prominence, and mobile readability.
Done
8
Redesigned the entire email follow-up automation sequence and nurture cadence
Conducted a comprehensive redesign of the complete email follow-up automation, including content, timing, conditional logic, and visual design. Restructured the nurture cadence to optimize engagement rates, reduce unsubscribes, and improve conversion from lead to qualified opportunity.
Done
9
Created #1 Re-engagement Email Campaign
Designed and built the first email in the re-engagement sequence targeting leads who dropped off after the initial nurture. The email re-introduces BSL with a fresh angle, highlights any new offerings or promotions, and provides an easy path to restart the conversation.
Done
10
Created #2 Re-engagement Email Campaign
Built the second re-engagement email, designed as a follow-up to Re-engagement Email #1 for contacts who opened but did not act. This email uses a different creative angle and stronger urgency-based messaging to drive re-engagement from dormant leads.
Done
11
Created SMS Snippet #1 for Lead Nurture Workflow
Wrote the first SMS message in the lead nurture sequence. The message is concise (under 160 characters), personalized with the contact's first name, and includes a clear value statement and call-to-action. Designed to generate an immediate reply or callback.
Done
12
Created SMS Snippet #2 for Lead Nurture Workflow
Wrote the second SMS message, sent 5 days after the first. This follow-up SMS uses a different approach (e.g., social proof, urgency, or question-based) to re-engage contacts who did not respond to the first text. Kept within character limits and compliance guidelines.
Done
13
Created SMS #1 for Lukewarm Lead sequence
Drafted a tailored SMS message specifically for the lukewarm lead segment. The messaging is softer and more educational than the hot lead SMS, acknowledging that the contact may still be researching options and positioning BSL as a helpful resource rather than a hard sell.
Done
14
Created SMS #2 for Lukewarm Lead sequence
Wrote the second lukewarm SMS, designed as a gentle follow-up that references a specific benefit or success metric to spark interest from contacts who did not respond to the first message.
Done
15
Updated SMS snippet per Slack thread feedback
Revised the SMS copy based on feedback received in the team Slack channel. Applied the suggested edits to tone, phrasing, and call-to-action language, then updated the live workflow with the approved version.
Done
16
Sent the lukewarm lead nurture flow and the two follow-up emails (including the TrustPilot request) to Slack for Carmine's review
Compiled the complete lukewarm lead nurture workflow, including all email templates, SMS messages, timing sequences, conditional logic, and the TrustPilot review request email. Posted the full package to Slack for Carmine's review and approval before activating the automated sends.
Done
Tracking, Attribution & Conversion
Pixel installation, conversion tracking setup, UTM attribution, click-fraud protection, and cross-platform tracking infrastructure
13
1
Installed Google Tag Manager (GTM) on the website
Deployed Google Tag Manager across the entire Bay Street Lending website, establishing the tag management foundation for all subsequent tracking implementations. GTM enables centralized management of all marketing pixels, conversion tags, and analytics scripts without requiring direct code changes to the website for each new tag.
Done
2
Set up the Meta Pixel (sent instructions to Vadym)
Configured the Meta (Facebook) Pixel for the BSL website to track conversions from Meta ad campaigns. The pixel captures page views, lead form submissions, and other key events. Sent implementation instructions to Vadym for installation on the website. The pixel enables Meta's algorithm to optimize ad delivery based on actual conversion data.
Done
3
Set up LinkedIn "Submit Lead Form" conversion tracking
Configured conversion tracking on the LinkedIn Ads platform specifically for lead form submissions. This enables LinkedIn's bidding algorithm to optimize for actual lead submissions rather than just clicks, and provides accurate conversion reporting in the LinkedIn campaign manager.
Done
4
Completed Bing Ads domain verification for conversion tracking
Verified the Bay Street Lending domain within the Microsoft/Bing Ads platform, a required step before conversion tracking can be activated. Domain verification confirms ownership of the website and enables the UET (Universal Event Tracking) tag to fire properly and send conversion data back to the Bing Ads platform.
Done
5
Created and tested conversion tracking on the Bing Ads account
Built the full conversion tracking infrastructure for Bing/Microsoft Ads: created the UET tag, deployed it via GTM, defined the conversion goal (lead form submission), set the conversion window, and ran end-to-end tests to verify conversions are firing and reporting correctly in the Bing Ads dashboard.
Done
6
Created and tested a custom "leadform" event for Reddit Ads conversion tracking
Built a custom conversion event called "leadform" in the Reddit Ads platform to track lead form submissions from Reddit ad campaigns. Deployed the Reddit Pixel, configured the custom event trigger, and validated the tracking with test submissions to ensure accurate conversion reporting.
Done
7
Linked the Bing Ads account to Looker for dashboard reporting
Connected the BSL Bing/Microsoft Ads account to the Looker reporting platform, enabling automated data pipelines that pull Bing ad performance data into centralized dashboards. This gives the BSL team a single view of ad performance across all platforms.
Done
8
Investigated and flagged a click-fraud pattern in Bing/Microsoft syndication network traffic
Discovered a significant click-fraud pattern originating from the Microsoft/Bing syndication partner network (search partner sites). The fraudulent traffic was generating fake form submissions that created bogus CRM entries, wasting ad spend and polluting lead data. Documented the pattern, flagged it to the team, and recommended immediate remediation steps including ClickCease deployment and syndication network exclusions.
Done
9
Added ClickCease click-fraud protection to the ad accounts
Deployed ClickCease across BSL's ad accounts to automatically detect and block fraudulent clicks, bot traffic, and competitor click activity. ClickCease monitors incoming ad clicks in real-time, identifies suspicious patterns (repeated clicks from same IP, data center traffic, click farms), and automatically adds offending IPs to the Google Ads and Bing Ads exclusion lists. This directly addresses the syndication network click-fraud issue identified during the audit.
Done
10
Followed up with Vadym about automated leads and adding qualifying questions to the website form
Coordinated with Vadym on the website lead form, recommending the addition of qualifying questions (loan amount, business revenue, time in business) that would help the sales team pre-qualify leads before outreach and improve lead quality scoring in Salesforce.
Done
11
Called/Slacked Vadym directly and reviewed AI notes on the last meeting
Direct coordination with Vadym to follow up on pending technical implementation items, review action items from the previous client meeting, and ensure alignment on integration timelines and deliverables.
Done
12
Set up Google Tag on the website
Configuration of the Google Ads conversion tracking tag (Google Tag / gtag.js) on the BSL website for accurate Google Ads conversion measurement.
Pending
13
Set a reminder to notify the team whenever a Google-triggered lead is observed, for attribution reconciliation in Salesforce
Due to limitations in Salesforce's native attribution for Google Ads leads, the team agreed to implement a manual notification protocol: whenever a Google-triggered lead is identified in the CRM, the team will be alerted so they can manually reconcile the attribution data, ensuring Google Ads gets proper credit for leads it generated.
Pending
Ad Platform Operations
Google Ads campaign management, AI automation enablement, and campaign QA
4
1
QA and launched Google Ads campaigns
Performed comprehensive quality assurance on all Google Ads campaigns before launch: verified targeting settings, geographic exclusions, bid strategies, conversion tracking configuration, ad copy compliance, landing page URLs, negative keyword lists, and budget allocation. After QA passed, activated the campaigns.
Done
2
Checked the Business Line of Credit campaign in Google Ads
Audited the performance and configuration of the Business Line of Credit Google Ads campaign. Reviewed keyword performance, search term reports, cost-per-acquisition, quality scores, and ad relevance. Identified optimization opportunities and made recommendations for improvement.
Done
3
Enabled AI Max text-asset automation and final URL expansion on "Search | SBA Loans | Government Backed" campaign
Activated Google's AI Max (formerly Auto-Applied Recommendations) text asset automation on the SBA Loans campaign. This feature allows Google's AI to dynamically generate and test additional headline and description variations. Also enabled final URL expansion, which allows Google to automatically direct users to the most relevant landing page on the BSL website based on their search query, rather than being locked to a single landing page URL.
Done
4
Enabled AI Max text-asset automation and final URL expansion on "Search | Industry Specific | Construction Trucking Restaurant" campaign
Applied the same AI Max text-asset automation and final URL expansion settings to the Industry Specific campaign targeting construction, trucking, and restaurant verticals. This gives Google's AI the flexibility to test new ad copy combinations and route users to the most relevant industry-specific landing pages automatically.
Done
AI Initiatives
AI chatbot development, AI-powered reporting, and experimental AI automation projects
4
1
Meeting participants tested the AI chatbot and provided feedback for iteration
Deployed an AI chatbot for BSL and conducted a live testing session with team members. Participants interacted with the chatbot, evaluated response quality, accuracy, and tone, and provided structured feedback. The feedback was compiled to guide the next iteration of chatbot training and prompt engineering.
Done
2
Looked into automating AI-driven Reddit commenting and reported back on findings
Researched the feasibility and compliance implications of using AI to automate responses and engagement in Reddit threads relevant to BSL's target market (small business lending, SBA loans, business financing). Evaluated available tools, Reddit's terms of service, and potential risks. Reported findings and recommendations to the team.
Done
3
Began planning to pull data through the OpenAI platform API to build a new AI-related reporting dashboard
Initiated the planning phase for a new reporting dashboard that leverages the OpenAI API to aggregate, analyze, and visualize BSL marketing data with AI-powered insights. The dashboard would provide automated performance summaries, anomaly detection, and recommendation generation.
In Progress
4
Explored Salesforce-to-HubSpot data transfer as part of CRM modernization
As part of evaluating modern CRM capabilities including AI-powered features, investigated the feasibility of migrating or syncing data between Salesforce and HubSpot. This exploration included assessing HubSpot's AI tools for lead scoring, predictive analytics, and automated content generation compared to Salesforce's native capabilities.
In Progress
Rebrand & Identity
Transition from Bay Street Lending to Bay Street Capital, including domain migration, logo updates, and email address rollover
5
1
Requested approval for rebranding to "Bay Street Capital" across ads, business listings, email domains, and all marketing assets
Formally proposed the rebrand from "Bay Street Lending" to "Bay Street Capital" across all customer-facing and internal marketing touchpoints. Outlined the scope of changes required: ad copy and creative updates, Google Business Profile rename, email domain migration, website copy changes, social media profile updates, and print/digital asset redesign. Submitted for Carmine's approval before initiating the transition.
Done
2
Updated the internal tools domain to BayStreetCap.com, while the public website stays on baystreetlending.com
Configured the new BayStreetCap.com domain for internal tools and CRM operations, while maintaining baystreetlending.com as the public-facing website domain during the transition period. This dual-domain approach ensures operational continuity while the rebrand rolls out in phases.
Done
3
Updated logos and business descriptions to read "Bay Street Capital, formerly Bay Street Lending"
Redesigned logo assets and updated all business descriptions across marketing platforms, ad accounts, and internal documents to reflect the new branding: "Bay Street Capital, formerly Bay Street Lending." The "formerly" designation ensures existing clients and leads recognize the continuity of the brand during the transition.
Done
4
Migrated roughly 90% of agents over to @baystreetcap.com email addresses
Executed the email migration for approximately 90% of BSL's agent team, transitioning their email addresses from the legacy @baystreetlending.com domain to the new @baystreetcap.com domain. This involved provisioning new mailboxes, configuring forwarding rules during the transition, updating email signatures, and ensuring CRM routing rules recognize both old and new addresses.
Done
5
Raised whether to formally kick off the broader marketing rebrand, given that 90% of agents had already been migrated
With the majority of the internal email migration complete, flagged to the BSL team that the operational groundwork was in place to formally launch the public-facing marketing rebrand. Recommended a coordinated rollout plan covering ads, social media, website, and business listings to ensure the transition is clean and unified rather than piecemeal.
Done
Client Communications & Reporting
Performance reports, material delivery, approval workflows, and stakeholder coordination
5
1
Delivered the Bay Street Lending Performance Report covering June 29 through July 12, 2026
Compiled and delivered a comprehensive performance report covering ad spend, lead volume, cost-per-lead, conversion rates, and platform-by-platform breakdowns for the June 29 through July 12 period. The report included insights on campaign performance trends, keyword analysis, and strategic recommendations for the upcoming period.
Done
2
Sent required material to Carmine
Prepared and delivered the requested marketing materials to Carmine for internal review. This included campaign assets, performance data, and strategic recommendations that required stakeholder approval before implementation.
Done
3
Sent the deck/materials required for Carmine's approval
Compiled a presentation deck and supporting materials for Carmine's review. The package included proposed campaign strategies, creative concepts, budget recommendations, and automation workflows that required executive sign-off before launch.
Done
4
Uploaded TrustPilot contact list to the Melleka App, tagged them, and scheduled the TrustPilot review email
Imported the verified list of funded contacts eligible for TrustPilot review requests into the Melleka App. Applied the "TrustPilot" tag to all contacts in the list, then configured and scheduled the review request email to send at the optimal time for maximum response rate.
Done
5
Retrieved the list of one-time TrustPilot contacts
Pulled the specific list of contacts who should receive a one-time TrustPilot review request (as opposed to being enrolled in an ongoing automation). These contacts were identified based on their funded status and TrustPilot eligibility criteria, and were separated from the recurring automation list for a single targeted send.
Done

Project Timeline

April 2026
✓ Zapped the Meta lead notification for instant sales team alerts
✓ Created Zap: Meta lead form submission to email notification
✓ Began auditing Salesforce automations and routing logic for leads
May 2026
✓ Installed Google Tag Manager (GTM) on the website
✓ Created Melleka App sub-account and invited BSL clients
✓ Integrated Meta on the Melleka App
✓ Purchased and verified toll-free number for marketing SMS
June 2026
✓ Connected Salesforce to the Melleka App
✓ Set up the Meta Pixel
✓ Created Zap: web form submissions to Melleka App
✓ Set up email domain for CRM sends
✓ Created Email Campaigns #1 and #2 for Lead Nurture
✓ Created SMS Snippets #1 and #2 for Lead Nurture
✓ Checked Meta leads being converted into website source
✓ Obtained backend access from Carmine
July 2026
✓ Delivered the Bay Street Lending Performance Report (Jun 29 - Jul 12)
✓ Mapped Facebook lead form fields into Salesforce fields
✓ Set up Salesforce Marketing Module access
✓ Built improved attribution logic in Salesforce
✓ Discussed email domain strategy for CRM/marketing sends
August 2026
✓ Created JustCall webhook workflow for missed calls
✓ Built Melleka App workflow: JustCall missed call to contact + tag + email
✓ Created Zap: Salesforce lead updates to Melleka App sync
✓ Audited all 54 Salesforce automations and email templates
✓ Built and deployed 4 lukewarm lead nurture emails (HTML-optimized)
✓ Created SMS #1 and #2 for lukewarm lead sequence
✓ Implemented Salesforce DND sync with Melleka App
✓ Updated workflow triggers and timing for lukewarm sequence
✓ LinkedIn Submit Lead Form conversion tracking setup
✓ Turned off PayPal Sent/Check Sent zap per Maria
✓ Sent lukewarm nurture flow to Carmine for review
✓ Created DND + internal notification workflow for stop replies
September 2026
✓ Created Zap: closed leads from Salesforce to Melleka App
✓ Created Re-engagement Email Campaigns #1 and #2
✓ Set up TrustPilot automation: Salesforce webhook + named credentials
✓ Configured TrustPilot triggers (funded amount > 0, exclusion = false)
✓ Uploaded TrustPilot contact list and scheduled review email
✓ Added ClickCease click-fraud protection
✓ Completed Bing domain verification and conversion tracking
✓ Created Reddit Ads custom "leadform" conversion event
✓ Linked Bing Ads to Looker for dashboard reporting
✓ Investigated and flagged Bing syndication click-fraud pattern
✓ Rebrand: Updated domains, logos, and migrated 90% of agent emails
✓ Enabled AI Max text-asset automation on 2 Google Ads campaigns
MELLEKA MARKETING
Prepared for Bay Street Lending / Bay Street Capital
April through September 2026
melleka.com