LendingUSA / Work at Bitcoin IRA
LendingUSA
Performance marketing for a multi-vertical point-of-sale lender
My role: paid media and demand generation for LendingUSA in 2017, part of my role as Director of Demand Generation at Bitcoin IRA.
$1.95M
Loan approvals, 316 in all
January 2017
69
Funded loans, $479.6K in principal
January 2017
$10K+
Past spend found on irrelevant keywords, then excluded
March 2017 audit
My role: paid media and demand generation for LendingUSA, within my role as Bitcoin IRA's Director of Demand Generation. Approvals and funded loans are company figures for the month, not revenue, and are not all attributed to paid media. The $10K+ is spend before the audit, not a monthly saving.
Overview
LendingUSA offered point-of-sale financing: installment loans that providers and merchants could offer their customers for dental and medical care, cosmetic procedures, fertility treatment, veterinary and pet care, funerals and other purchases. Its marketing had two audiences at once: consumers looking for financing, and the providers and merchants who wanted a financing program for their customers.
I started with a consulting audit in January 2017 and went on to run the paid search and display program: rebuilding the account by financing use, matching landing pages to intent, resolving platform policy issues, and building the tracking needed to judge campaigns on approvals and funded loans rather than clicks.
The challenge
A complex lending funnel with little visibility
The legacy Google AdWords and Bing accounts mixed general finance terms with a competitor campaign, carried broad and irrelevant keywords, and sent borrowers and merchants through overlapping paths. Some landing pages were missing Google Analytics tags, PPC codes were inconsistent, approvals and rejections shared one destination, and funding happened offline, so it was hard to tell which spend produced loans.
Lending is also a regulated category on ad platforms. Ad copy, sitelinks, call extensions and landing pages all had to meet Google's financial-services rules, or delivery could stop no matter how good the targeting was.
Goals and scope
Goals
- Cut spend on traffic that would never become an approved loan
- Match each search to the financing use and the right landing page
- Separate consumer borrowers from providers and merchants
- Keep ads and pages within platform financial-services policy
- Track each step from click to application, approval and funded loan
Scope
- Paid search
- Display and Gmail
- Remarketing
- Landing pages
- Ad copy and extensions
- Conversion tracking
- Daily reporting
Teams coordinated
- LendingUSA marketing
- Developers
- Google account support
The playbook
Rebuilding the account around loan intent
-
Audited the legacy accounts
I reviewed the Google AdWords and Bing accounts' search queries, ad groups, budgets and lead quality, and found more than $10,000 of past spend on keywords that had nothing to do with financing, which I then excluded. I paused broad, "cheap" and "bad credit" terms, tightened negative keywords, and adjusted mobile and income-segment bids.
-
Restructured campaigns by financing use
I moved general finance terms into their own campaign, apart from the competitor campaign, whose bids and budget came down after costly traffic. My new ad groups followed the procedure or purchase being financed: dental, cosmetic and bariatric procedures, fertility and IVF, chiropractic care, funerals, pet and veterinary care, and general merchant financing, with legal, wedding and plastic surgery campaigns added in the spring.
-
Split borrowers from providers
Each financing category got a consumer application path and a separate page for providers and merchants requesting a financing program, with sitelinks for each kind of visitor. I built or reorganized search, display, Gmail and remarketing campaigns around those two journeys, with consistent UTM conventions across the sites.
-
Resolved policy issues with Google
When Google suspended the account in April 2017 over financial-services transparency, promotional claims and sensitive procedure copy, I removed or paused the flagged legacy ads and landing pages and submitted an appeal, working with Google and LendingUSA contacts. In May, I fixed call extensions so their phone numbers matched their landing pages.
-
Built the measurement
I diagnosed missing analytics tags, inconsistent PPC codes and a shared approval and rejection page with the developers, who added Google Tag Manager and application events. My daily reports and campaign reviews then compared spend with applications and approvals where the data allowed, so budget decisions rested on loans, not click-through rates.
Key moves
What changed
- Irrelevant, broad and "bad credit" terms cut, with tighter negative keywords
- General finance separated from the competitor campaign
- Ad groups built around each procedure or purchase being financed
- Separate journeys for borrowers and for providers and merchants
- Flagged ads and pages removed and an account suspension appealed
- Tag Manager, application events and UTM conventions across the sites
Results
The numbers
Approvals and funded loans are different measures, and neither is lender revenue.
$1.95M
Approved loan amounts, 316 approvals
January 2017, company figure
$479.6K
Funded principal across 69 loans
January 2017, company figure
$10K+
Past spend on irrelevant keywords, found and excluded
March 2017 audit
Results in detail
What the numbers mean
The $1.95M is the total of loans approved in January 2017; the $479.6K is what was actually lent across the 69 loans funded that month. An approval is not a funded loan, and neither is revenue for the lender.
Funding happened offline and some pages were still missing tracking, so the records do not show every approved or funded loan coming from the paid campaigns. They are company figures for the month, shown for scale.
The $10K+ is spend that went to irrelevant keywords before the audit. Excluding them stopped that waste going forward, but it is not a measured monthly saving.
Timeline
2017, from audit to a structured account
January 2017
I began consulting with an audit of the paid search accounts.
February 2017
A new branded landing page went live, with tracked clicks to the credit application.
March 2017
$10K+ of past spend on irrelevant keywords found; general finance split from the competitor campaign.
April 2017
Legal, wedding and plastic surgery campaigns added; flagged ads removed and a Google suspension appealed.
May 2017
Call extensions matched to their landing pages' phone numbers.
July and August 2017
Segmented remarketing, device bids, consumer sitelinks and a separate fertility and IVF structure.
Key learnings
Learning Experiences
- Judge keywords by approved and funded loans, not click-through rate. An irrelevant term can have a high click-through rate and never produce a loan.
- Keep branded, competitor, specialty and merchant campaigns separate enough to control bids, message, landing pages and compliance review for each.
- Track every step on its own: click, landing page action, application, approval and funded loan. A form submit is not a funded loan.
- Review the landing page, disclosures and phone numbers with the ad. In financial services, a policy issue can stop delivery even when targeting is right.
FAQ
Questions about this case study
How much of the $1.95M came from paid media?
The records do not say. Funding happened offline and tracking was still being built, so the approvals and funded loans are company figures for January 2017, not results attributed to the campaigns.
What is the difference between an approval and a funded loan?
An approval means the borrower qualified for an amount. A funded loan means the money was actually lent. Many approvals never fund, which is why the page shows both.
What did I do, and what did the company do?
I audited and rebuilt the paid search and display accounts, set up the borrower and provider journeys, handled the policy appeal, and worked with developers on tracking. LendingUSA's own teams handled underwriting, funding, and merchant relationships. The approvals and funded loans are company figures for the month, not results attributed to my campaigns.
Contact
Let's connect.
Questions about this work, a role, or a project of your own? I read every message myself.