1.4x improvement in response rates and 2x increase in campaign efficiency
A Fortune 10 retailer was running high-volume direct mail campaigns with a fixed approach to offer assignment. AIgnyte matched every prospect to their individually optimal offer combination — selecting from 4,000+ possibilities — with no new creatives and no changes to the existing execution stack.
The right question was being overlooked.
This Fortune 10 retailer ran extensive direct mail campaigns — large files, tested offers, optimised targeting. The existing approach focused on a well-established question: who is likely to respond? Response models, selection criteria, and segment-level targeting were all built around that framing.
What wasn't being asked — and what no existing system could answer — was a different question: what will stimulate this specific prospect to respond? The retailer had a deep offer catalog. Multiple product combinations, price points, promotional structures, and message framings were all in play. But every prospect in the file received the same offer, regardless of which of the thousands of possible combinations was most likely to move them.
The untapped value wasn't in the targeting. It was in the offer matching. And it was large.
Three stages. 4,000+ combinations. Every individual.
AIgnyte's standard three-stage methodology: every offer combination in the retailer's catalog encoded into a Genetic Signature; at campaign time, a fresh Preference Profile derived for every prospect — i.e. how will each offer element influence this individual to respond or not, inferred from their own attributes and how others like them have responded and not responded before; then each prospect matched to the highest-scoring offer from 4,000+ possible combinations. One individually-reasoned offer decision per person, every campaign cycle. Output delivered in the client's existing file format — no changes to their execution stack.
1.4x response rate. 2x campaign efficiency. Same catalog, same data, same process.
- Reframing the problem from who will respond to what will make each individual respond
- Matching each individual to the optimal offer from 4,000+ combinations — not the best on average
- 2x campaign efficiency — more response per pound of campaign spend, with the same budget
The value was already in the catalog.
The retailer's existing models were optimised to answer who is likely to respond. AIgnyte added the second layer — given that this person is being mailed, what offer is most likely to make them respond? These are separate problems. With 4,000+ possible combinations and millions of prospects, no manual process could answer the second question at scale. AIgnyte did — making a distinct, individually-reasoned offer decision for every person in the file, entirely complementary to the existing targeting approach, without touching the selection process or the execution stack.
Your offer catalog is more valuable than you think.
AIgnyte works with your existing offers, your existing campaign file, and your existing execution stack. The lift comes from matching — not from building more.