Eximpedia · B2B SaaS · Trade intelligence
The data was never the problem. The reach was.
- Client
- Eximpedia
- Sector
- B2B SaaS · Trade intelligence
- Market
- Singapore, global reach
- Engagement
- 12-month growth retainer
- Practices
- Search · Paid · Content · Lifecycle
2×
website visitors in 12 months
+40%
qualified demo requests
−20%
cost per qualified lead
60%
non-brand search entries
The problem
A trade-intelligence platform good enough for enterprise research teams, found mostly by accident. Buyers who needed exactly this product were arriving late or not at all, and the traffic that did arrive was too broad to convert — one product being sold to four different buyers with a single message.
What we did
- Spent the first two weeks on diagnosis rather than campaigns, and found one product addressed to four distinct buyer groups with one undifferentiated message.
- Rebuilt search and content around the one asset no competitor could copy: Eximpedia’s own shipment records, in a repeatable answer-first format with sourced tables, explicit HS codes and named bylines.
- Rebuilt Google Ads around commodity and country-pair intent, ran LinkedIn against role and firm type, and used Meta for retargeting and the owner-operator exporter segment.
- Gave each segment its own landing destination and its own offer, so the click and the page it landed on agreed with each other.
- Built email as infrastructure rather than a newsletter: trial-to-paid onboarding, lapsed-account re-engagement, and a monthly trade-movement brief that holds customers between renewal decisions.
Eximpedia sells access to verified import–export shipment records — buyers, suppliers, HS codes, ports, volumes, unit values — across more than two hundred countries, through a dashboard built for people who make sourcing and market-entry decisions for a living.
The product had already earned its way into research teams at large consultancies and financial institutions. That is a hard bar to clear, and it had been cleared. What had not been solved was distribution. Buyers who needed exactly this platform were finding it late, or not at all, and the traffic that did arrive was too broad to convert.
The brief we were given was short: expand outreach to the right customer base, and reach global markets, quickly.
What we found when we looked
We spent the first two weeks on diagnosis rather than campaigns. Three things surfaced.
One product, four buyers, one message
Eximpedia serves at least four distinct groups: exporters and manufacturers hunting buyers; logistics providers; financial institutions modelling supply-chain risk; and consulting firms building market reports. Each buys for a different reason, on a different cycle, with a different definition of value. The site addressed them with a single message aimed at everyone, which in practice landed with no one strongly enough to convert.
Content that could have been written by any competitor
The blog covered the right topics and ranked for very little. The reason was structural: the posts read like general trade explainers. They did not use the one asset no competitor could copy: Eximpedia’s own shipment records. A hundred near-identical articles about coffee exports already exist. One with real, cited, multi-year figures attached does not.
Paid spend without a segment behind it
Advertising was reaching volume, not fit. Without a mapping from segment to channel to offer, spend optimised toward the cheapest click rather than the most valuable conversation.
The working thesis
A trade-data platform does not have a traffic problem. It has a specificity problem. Every buyer arrives with a named commodity, a named country pair, or a named HS code in their head. Meet them at that level of specificity and the conversion work is already half done.
The plan
Five channels, run as one system off a single segment map, not five agencies pulling in five directions. Every channel had one job, and the jobs were sequenced.
| Channel | Job it was hired to do |
|---|---|
| Search & content | Own the specific, high-intent queries: commodity + country + data. Compound over months, cost nothing per click. |
| Google Ads | Cover the same intent immediately while the content compounds. Buy the terms we had not yet earned. |
| Reach the four segments by role and firm type — the only channel where “trade analyst at a logistics firm” is directly addressable. | |
| Meta | Cheap reach for retargeting and for the owner-operator exporter segment, which does not live on LinkedIn. |
| Convert the traffic the other four channels created, and hold the customers already won. |
What we actually did
Search and content, rebuilt around proprietary data
We replaced generic explainers with a repeatable format: an answer-first opening containing the headline figure, a findings list near the top, headers that carry the finding rather than a bare label, at least two data tables with source and period captions, HS codes stated explicitly with their product description, and a numbered methodology note before the FAQs. Every post is bylined by a named person with a real role.
That last set of details is not house style for its own sake. It is what makes a page mentionable by an answer engine deciding which source to cite. Structured, sourced, attributable content is the format AI search reaches for.
Paid, mapped segment by segment
Google campaigns were rebuilt around commodity and country-pair intent rather than broad category terms. LinkedIn ran segment-specific creative against role and firm-type audiences. Meta carried retargeting and the exporter segment. Each segment had its own landing destination and its own offer: a sample report for the researcher, a live dashboard walkthrough for the analyst.
Email as infrastructure, not a newsletter
Three sequences: a trial-to-paid onboarding flow, a re-engagement flow for lapsed accounts, and a monthly trade-movement brief that gives existing subscribers a reason to open the dashboard. The brief doubles as the brand-building programme: it keeps Eximpedia in front of customers between renewal decisions, which is the moment retention is actually won.
What changed
Website visitors doubled over 12 months — and, more usefully, the composition of that traffic changed. 60% of sessions now arrive on commodity- or country-specific pages, which is traffic that arrives already knowing what it wants.
Qualified demo requests increased by 40%. Cost per qualified lead fell 20% as paid spend shifted from broad terms to segment-mapped campaigns.
What we have done
We ran paid and content in parallel from week three. In hindsight, paid should have led by four to six weeks on its own. Search queries that convert on Google Ads are the cheapest possible research for deciding which content to write next, and we spent the first month writing on informed guesses instead of on paid conversion data we could have had.
We also underweighted the financial-institution segment early. It has the longest sales cycle of the four and the highest contract value, and long-cycle segments look like failures on a 30-day dashboard. They need to be measured on a different zone, and we should have set that expectation at the start rather than defending it at month two.
Same problem, different product?
If you have something good that the right people cannot find, send us a brief (or half of one). You get scope, timeline and a number back within three working days.
Published with the client's review and permission. More work
Next step
Ready when you have a brief.Or half of one.
Send whatever you have. You get scope, a timeline and a number back within three working days.
Prefer chat? We answer on WhatsApp too.