Salesforce wants Braze. Customer data matters when a business can act on it

An online store may know exactly how many people viewed a product on Tuesday and still lose track of the person who asked about availability on Wednesday. Traffic analytics record visits. Campaign dashboards record acquisition costs. Sales reports record orders. The customer’s journey between those systems is often far less visible.

On October 20, 2026, Salesforce announced a definitive agreement to acquire Braze at an approximate enterprise value of $5.66 billion. Braze helps businesses respond to behavioral signals across websites, apps, email, SMS and WhatsApp. The deal remains subject to shareholder and regulatory approvals, with closing expected during Salesforce’s fiscal 2028. It has not closed, and proposed product combinations should not be described as existing integrations.

The more useful business question sits beneath the transaction price. What happens when a company recognizes a customer’s intent but cannot act on it before the context disappears?

A visit is not a customer relationship

Anonymous browsing data can reveal patterns without identifying the person behind them. It does not create permission to send marketing messages. The relationship changes when someone requests a quote, signs up for a specific alert or otherwise supplies information for a defined purpose. At that point, a useful record needs more than a name and an email address. It needs the request, its context, the permission granted and the outcome.

Consider a hypothetical premium footwear retailer. A shopper asks to be notified when a particular size returns. The product becomes available the following week. If that request sits in an isolated form inbox, the inventory team may have the shoes while the customer hears nothing. If the request is connected to the product, stock status and the shopper’s consent, the retailer can send a relevant update. If the customer has already bought the shoes, the system should know that too.

This is not primarily a question of which CRM has the longest feature list. The failure can occur at any handoff: between a form and inventory, between a salesperson and the shared customer record, or between a purchase event and a scheduled campaign. A perfectly functioning website can still produce a fragmented customer experience.

The six links between a signal and a useful response

A practical review begins with the signal: a product inquiry, booking request, abandoned application or availability alert. Next comes permission: what information can be retained, and for which purpose? Identity follows: which records genuinely belong to the same customer, without overconfident assumptions about devices or accounts? Then comes the response: what would help, through which channel, at what frequency? Finally, the business needs an outcome and a measurement method that distinguishes observed results from assumptions.

Signal → permission → identity → response → conversion → measurement. This is an editorial decision framework, not a product architecture announced by Salesforce.

Consent has to survive the journey between systems. Permission to receive one stock alert is not blanket permission for recurring promotional messages. A website visit alone is not consent. Identity matching should also be treated as a fallible process rather than an excuse to merge unrelated profiles.

What Salesforce is buying into

Salesforce says Braze’s real-time engagement capabilities are expected to complement its customer data, CRM and agent products following the proposed acquisition. The strategic interest is understandable: customer information becomes more useful when it can inform a timely, appropriate action.

That does not mean a combined technology stack will automatically improve outcomes. Data can be duplicated, consent can be recorded inconsistently, and teams can continue using workflows that never close the loop after a sale. Those problems require operating decisions as well as software. The transaction itself is still pending.

For an organization already deeply invested in Salesforce, the eventual product direction may deserve close attention. For a business using smaller tools, connecting its existing storefront, customer records and communication channels may be the better investment. Neither answer should be assumed from the size of an acquisition.

Owned channels need operational memory

Brands that rely on paid platforms to reacquire the same people repeatedly may be paying for demand they had already earned. An owned channel can support an ongoing, permission-based relationship, but only when customer interactions retain their context. A website and a contact database are not enough on their own.

The connection between digital architecture and conversion is therefore operational, not merely visual. A form that is easy to complete but never reaches the person responsible has not finished its job. That is why work on owned digital channels and their architecture needs to consider what happens after the visitor clicks Submit.

The issue also intersects with purchases moving into third-party interfaces. Our earlier examination of checkout inside conversations explored how platforms can control the transaction surface. The Salesforce–Braze announcement raises the next question: once a customer has interacted with a brand, what context can the business legitimately retain and use?

What to inspect before choosing another platform

Where does a request go when it arrives through the website, WhatsApp, email or a salesperson? If every channel produces a separate record, the company may have plenty of activity and very little shared customer knowledge.

Who checks consent before an automated message goes out? Personalization without a verifiable permission rule is a liability, not a lasting advantage.

And who records when the customer has already received an answer or completed the purchase? Without that feedback, automation can become repetition, and performance reports can overstate what a campaign actually achieved.

Salesforce has announced a multibillion-dollar plan to bring customer signals and engagement technology closer together. A business does not need a transaction of that size to test the underlying idea. It needs to follow one real customer request from the first signal to the final outcome and identify where the context gets lost.