Start with the Top of Funnel: Sign-up Forms
Whenever I open a Klaviyo account, I work from the top of the funnel straight down to the bottom. That means sign-up forms come first. For an 8-figure brand doing around $1.2 million every 30 days, sign-up forms are the lifeblood of subscriber acquisition.
The first rule is simple: separate your forms by device. You should never run a single form that caters to both desktop and mobile. If you want true A/B testing data, you must isolate them. Variation A might win on desktop, but variation B could convert far better on mobile. When they are combined into one form, you cannot analyse how different devices interact with your offer.
Next, look at the submission flow and zero-party data capture. I like to ask interest-based questions (such as who they are shopping for or what categories they prefer) right on step two. Most importantly, always provide the discount code on the success step of the form. Forcing a prospective customer to leave your site, open their inbox, and hunt for a coupon code creates unnecessary friction. That is a mistake.
Finally, never capture SMS on the same form as email. When brands force or even prompt for a phone number on the initial pop-up, 25% to 45% of users simply close the form. That means nearly half your traffic never sees the welcome discount on site. Keep email capture and SMS capture completely separate.
Account Settings and Technical Hygiene
Once the forms are validated, I dive into backend account settings. This is where basic deliverability and tracking flaws hide in plain sight.
First, verify that the dedicated sending domain is connected and authenticated properly with clean SPF, DKIM, and DMARC records. Next, check the data tracking settings. You want to make sure anonymous visitor tracking is enabled so Klaviyo can tie pre-identification browsing behavior back to customer profiles once they submit a form.
Look at your opt-in process. Nine times out of ten, you want single opt-in enabled across your lists. Double opt-in adds unnecessary drop-off unless your account is suffering from severe deliverability issues or spam bot attacks.
Lastly, check attribution settings. Klaviyo defaults to a 5-day attribution window. I frequently see agencies artificially inflate this window to 20 or 30 days to make their reported revenue look massive. That is dishonest reporting. Keep your attribution tight so you are looking at real, incremented numbers.
Auditing the Core 12 Automated Flows
When auditing automations, filter by live status. Every mature ecommerce brand should have at least 12 foundational flows operating concurrently in their account.
- Welcome Series (Retail and Wholesale separated)
- Browse Abandonment
- Abandoned Cart
- Abandoned Checkout
- Post-Purchase Thank You
- Fulfilled Order
- Delivered Order
- Cancelled Order
- Refunded Order
- Customer Winback
- Sunset Unengaged
- Review Request
When evaluating live flows, check triggers and split logic. For browse and cart abandonments, ensure you have trigger splits separating existing customers from non-purchasers. A first-time shopper needs trust-building and an introductory incentive, while an existing buyer simply needs a gentle reminder.
Look for missed opportunities like category splits. If someone adds a specific product line to their cart, your automation should serve dynamic content and messaging tailored to that exact collection. I also check whether SMS messaging is interlaced into these core flows. If a brand is generating 32% of its revenue from Klaviyo but SMS is completely absent across the flow library, that is low-hanging fruit waiting to be claimed.
Campaign Cadence and List Segmentation
After automations, I review the campaign calendar and segmentation standards. For an 8-figure store, sending one campaign every two days is standard, but you must maintain a baseline of at least two campaigns per week.
Open rates and click rates give you an immediate snapshot of deliverability health, but the real test is how your engaged segments are constructed. You should never blast your entire database. Build a clean 90-day engaged segment that captures subscribers who have opened an email, clicked an email, visited the website, or placed an order within the last 90 days.
Split your campaign sends between engaged customers and engaged non-purchasers. The non-purchasers should receive campaign variations featuring the welcome offer to convert them into first-time buyers.
The reason is simple: roughly 80% of your email revenue will come from people who have already purchased once. On a recent campaign send I audited, the customer segment generated $7,262 while the non-purchaser segment brought in $640 from the exact same send. The money is in the repeat order. Use campaigns to push first-time buyers over the line, then let your product experience do the heavy lifting.
Turning Audit Findings into an Actionable Roadmap
Collecting data during an audit is only half the battle. You need to translate those technical notes into a visual roadmap that prioritises the highest-impact fixes first.
I map every audit finding onto a visual Miro board rather than burying it inside a slide deck. This allows you to visually lay out the missing automations (such as sunset flows, transactional follow-ups, and SMS steps), outline updated flow architectures, and map out upcoming segmentation rules side by side.
A clear audit should show exactly what is missing, what is broken, and what needs immediate optimisation. When you can see every gap across forms, settings, automations, and campaigns in one place, the execution plan becomes obvious.
Final Thoughts
Auditing an enterprise Klaviyo account is about finding the compound gains across every touchpoint. Tightening sign-up forms, fixing attribution, splitting automations by customer status, and segmenting campaign sends will radically shift your total revenue share. Small technical fixes turn into significant revenue gains when applied at scale.
If you want our team to review your account and build a custom growth roadmap for your store, request our account audit today.
