22. Deal story templates

Stories beat feature lists. Tell three workflow stories, not thirty features.

The template (use for every win, loss, and public case study)

1. Who: industry, size, persona who championed.
2. Before: where data lived, what hurt, with a number.
3. Trigger: why now.
4. Alternatives: who else they evaluated.
5. What we drew: the architecture in one sentence.
6. Proof: the POC criterion that decided it.
7. After: outcome with a number and a quote.
8. Lesson: what the next SE should copy.

Keep it to 150 words spoken, 90 seconds.

Public stories rewritten in the template (from backblaze.com case studies, checked 2026-10-14)

Decart (AI)

• Who: AI startup focused on real-time inference; CEO Dean Leitersdorf.
• Before: burning free credits at a traditional cloud; data growing from hundreds of thousands to hundreds of millions of video hours.
• Trigger: credits and scale.
• What we drew: B2 as the single store feeding GPU clusters around the world; Overdrive.
• After: 0 to 16 PB in 90 days; “10x more efficient than competition”; $0 egress spend.
• Quote: “B2 Cloud Storage was literally the only technical thing we used in training these models that didn’t crash the first time we tried it.”
• Lesson: lead with reliability under scale, not just price.

Hume AI (AI)

• Before: audio data across multiple hyperscale object stores past PB scale.
• What we drew: one unified object store with WEKA in the stack.
• After: close to 10 GB/s upload from clusters; $0 egress fees; one store.
• Quote: “Our growth doesn’t follow a curve, it follows a trigger.”
• Lesson: consolidation is a story on its own.

Mirage (AI video; URL slug “captions”)

• Before: high egress and inflexible regions at prior provider while moving to H100 clusters.
• Proof: competitive bakeoff.
• After: 0 GPUs waiting for data, 95% estimated cost savings, $0 egress (Overdrive).
• Lesson: run the bakeoff; Overdrive wins on dedicated throughput plus egress.

Philadelphia Eagles (sports media)

• Before: failure-prone LTO-6, slow access, no geographic redundancy.
• What we drew: Mimir MAM + Quantum QXS + B2 with Cloud Replication; UDM.
• After: 800 TB under management, LTO retired, 100% geographic redundancy.
• Quote: “We went from frequent LTO crashes and long restore times to near-instant access for every stored clip.”

Austin City Limits (public TV)

• Before: 4,000+ hours of deteriorating videotape.
• What we drew: Fireball + UDM into B2.
• After: 40+ years preserved; 40 TB transferred without delay.

AcenTek (telecom IT)

• Trigger: cyber insurance requirement.
• After: proved immutable off-site backups to insurer.
• Quote: “Adding Backblaze to our infrastructure allowed us to satisfy our insurance carrier’s requirements.”
• Lesson: insurance is the forcing function for backup deals.

Tribute (SaaS)

• After: $15K saved per month. “Backblaze was a drop-in API replacement with a much better cost structure.”

Pittsburg State University (education)

• After: off-site, ransomware-protected backups with Object Lock; “4x data protection”, “5 decades of data”.

Discovery-to-story worksheet (fill after each call)

Telling it

• Open with the “before” number.
• Pause after the trigger.
• Draw the architecture in five strokes.
• Close with the “after” number and ask: “Does that look like your situation?”