Case studies / Otrovato
Otrovato · Cross-border sourcing · Kenya · Client since Dec 2023
How Otrovato opened 49,518 WhatsApp conversations in Kenya, at half the cost it started with
A sourcing business selling from Egypt into Kenya, with no local salesforce and a category that does not buy from a web form. Eleven months of reach and traffic had produced 22 conversations. We moved the conversion point into WhatsApp and rebuilt the account around one number: the cost of a conversation.
Last updated 12 Sep 2026 · figures through Sep 2026 · source: Meta Ads Manager
49,518Conversations startedWhatsApp and Messenger, Dec 2023 – Sep 2026
3.7×Conversation volume, year on yearLast 12 months against the 12 months before
−48%Cost per conversationSame comparison, while media nearly doubled
The case in 30 seconds
The challenge
A category that negotiates before it buys, no local salesforce in the market, and an ad account measured in reach: eleven months of it produced 22 conversations.
What we changed
WhatsApp became the destination instead of the website, the account was rebuilt as reach → engagement → message, B2B and B2C were separated, and cost per conversation became the only scoreboard.
The result
49,518 conversations in Kenya, 3.7× the volume year on year at 48% less per conversation, and a quarter that ran 4.6× the first messaging quarter at 97% of its cost.
Our diagnosis
This category does not buy from a website. It buys in a chat.
One map of how the sale actually happens before rebuilding the account.
- Reach with no destinationEleven months of awareness and traffic, 22 conversations
- No local salesforceAn Egyptian team, Kenyan buyers, no branch to walk into
- A negotiated saleQuantity, price and delivery are agreed, not clicked
- Two audiences, one accountWholesalers and consumers wanted different things
- Spend without a signalSearch and shopping counted page views, not buyers
The opportunity
WhatsApp as the destinationThree-stage funnelB2B and B2C splitCost per conversation
→A pipeline the sales team can answerBefore → After
What actually changed
BeforeAfter
BeforeThe website as the destination
AfterWhatsApp as the destination: the chat is where the deal is agreed
BeforeReach and engagement as the result
AfterConversations started as the result, costed every week
BeforeOne funnel for everyone
AfterSeparate B2B and B2C message campaigns, each with its own creative and budget
BeforeTwo platforms, one of them unmeasured
AfterOne platform and one measured action; the unmeasured spend was closed in June 2025
BeforeBudget spread across markets
AfterKenya concentrated: 99% of the last twelve months in one market
Growth strategy
One growth system. Four connected levers.
01 — Media
Three stages, each funded for one job, rebalanced weekly on the cost of a conversation.
Reach→Engagement→Message→Sale
Reach and engagement build the audience; only the message campaigns are judged on cost.
02 — Creative
Product-and-price video that ends in one instruction: message us.
Seasonal pushes ran as separate campaigns so the always-on engine was never disturbed by a spike.
03 — Audience
Wholesalers and consumers separated at campaign level, not inside one ad set.
Case quantities and margin language on one side, single units and convenience on the other.
04 — Measurement
One number governs the account: cost per conversation started.
Reach and engagement inform the mix; the conversation decides what scales.
Performance over time
What happened next
Meta Ads Manager data by quarter. The first full messaging quarter, Q1 2025, is the benchmark: cost rose while volume scaled, then came back below the benchmark with 4.6× the volume. Q3 2026 is partial, to 6 September.
The business impact
Account numbers for the periods stated, Kenya.
49,518Conversations startedDec 2023 – Sep 2026
39,001In the last 12 months aloneSep 2025 – Sep 2026
3.7×Volume, year on year10,496 → 39,001
−48%Cost per conversationYear on year, media up 93%
4.6×Best quarter vs the firstQ2 2026 vs Q1 2025, at 97% of its cost
105MImpressionsOne market, 33 months
Media spend and revenue are confidential to the client. Figures are as reported in Meta Ads Manager for the periods stated; volumes, ratios and index values are shown, budgets are not. A conversation is a WhatsApp or Messenger conversation started from an ad; orders are agreed in the client's own inbox, which no ad platform can see.
Creative strategy
The creative system
The pieces that did the work, not the whole gallery.
- Format
- Product-and-price video · BOFU
Why it workedOne product, the wholesale price, and a single instruction to message. The highest-volume format in the account across two years.
- Format
- Seasonal push · Campaign
Why it workedBlack Friday, Valentine and flash sales ran in their own campaigns with their own budgets, so a spike never disturbed the always-on engine.
- Format
- B2B and B2C pair · Audience
Why it workedThe same product in two campaigns: case quantities and margin for wholesalers, single units and convenience for consumers.
What we learned
01
Reach is not a funnel without a destination.
Eleven months of awareness and traffic produced 22 conversations. The same market, pointed at a chat window instead of a web page, produced thousands a month.
02
Scale is proved by the cost holding, not by the volume rising.
Quarterly conversations reached 4.6× the first messaging quarter while the cost of a conversation came back to 97% of it. Volume on its own would have proved nothing.
03
This case stops at the conversation, and says so.
Orders are agreed in the client's WhatsApp inbox, outside any ad platform. Two markets tested outside Kenya produced 37 conversations between them and were closed, and one month ran at thirty times the normal cost before it was caught. Both are on this page for the same reason.
Generating leads but not enough of the right ones?
We rebuild the funnel around lead quality and cost per qualified action, not form fills.
