← The Journal
revops

Customers Log In Every Week—and Still Run the Real Process in Spreadsheets

Weekly logins can hide a customer workflow still running in spreadsheets. Learn how to trace the real job, classify workarounds, and measure end-to-end SaaS adoption.

By · 6 min read · 9 November 2026

A laptop workflow dashboard connects to a longer sequence of paper records on a spreadsheet ledger, showing work continuing outside the SaaS product.

Weekly logins tell you that customers return. They do not tell you whether your product has become part of the work customers bought it to do. If the important task still ends in an export, a spreadsheet, or a second round of manual reconciliation, you may have activity without end-to-end workflow adoption.

That gap is easy to miss in a healthy-looking usage dashboard. It is also easy to misread: a spreadsheet is not automatically evidence that your product failed. It is a clue to follow through the customer’s actual process.

Engagement shows a return visit. Adoption shows a job getting done.

A login is an event. A workflow is a chain: a trigger arrives, the right information is available, someone makes a decision, the work moves to the next person, and an outcome is recorded. A customer can log in every Monday and still rely on a different tool for the part of that chain that matters most.

Product-adoption guidance already points beyond raw visits. Amplitude’s overview of product-adoption metrics covers feature usage, activation, adoption rate, and time to activate. Gainsight’s lifecycle guide distinguishes onboarding measures such as time to value from feature adoption in mature use. Those metrics help teams see whether people return and which parts of the product they use. For an established account, add one more question: does the customer complete the recurring job in the product, end to end, with the people and data that job requires?

That is a useful working definition of workflow adoption—not a universal industry standard. It shifts the conversation from “Did someone use the product?” to “Where did the work finish, and what still had to happen elsewhere?”

Treat the spreadsheet as a clue, not a verdict

Imagine a customer who opens an analytics platform each week to check account activity. At month-end, an operations manager exports several reports, joins them in a spreadsheet, applies a rule the product does not support, and sends a version to a team lead for approval. The platform is in use. But the spreadsheet may still be where the customer’s actual review process is assembled and completed.

That example is illustrative, not a customer case. The distinction matters because the same workaround can point to different problems:

What you observe What to investigate
Data is exported, cleaned, and re-entered Is a required field missing, stale, or difficult to access? Is an integration or mapping absent?
A spreadsheet applies a rule or calculation Is the product missing a capability, or does the customer need a policy the product should not encode?
One person uses the product, then sends a file to colleagues Does the workflow depend on roles, permissions, approvals, or handoffs the product does not support?
A spreadsheet handles only exceptions Is the standard path working while a legitimate edge case needs a separate route?
The same information is entered in two places Who owns the handoff, and which system is meant to be the source of truth?

The answer may be a product gap. It may instead be a data dependency, a permission rule, a missing owner, a training issue, or a sensible customer control. Do not promise that every spreadsheet can—or should—disappear. First find out what job it is doing.

Trace the work, not just the clicks

A short workflow review can reveal more than another broad usage chart. Pick one recurring customer job and follow its most recent real instance from start to finish.

  1. Name the job and outcome. Write down what the customer is trying to achieve in plain language: for example, “review the accounts that need attention and assign the next action.” Define what counts as a completed result before choosing product events to measure.
  2. Ask the customer to walk through the last instance. With their consent, have the people involved show the sequence they actually followed—not the process in a training deck. Ask where they switched tools, copied data, waited for approval, or made a decision outside the product.
  3. Map the people, information, and handoffs. Note who starts the work, who needs to act, what data each person requires, and where responsibility changes. A workflow can appear successful in an administrator’s activity while remaining incomplete for the account team that owns the outcome.
  4. Compare the map with product telemetry. Look for the product events that correspond to meaningful steps, not just visits or page views. Segment by account, workflow, and role where possible. Product analytics can show what happened inside the product; it will not automatically reveal what someone did in a spreadsheet afterward.
  5. Classify the point where work leaves the product. Record the reason, the workaround, how often it occurs, who maintains it, and what risk or effort it creates. Validate the explanation with the customer before turning it into a roadmap request.

If the cause is an integration or a prerequisite, Rothenhall’s software integration checklist for buyers offers a useful way to examine supported data, permissions, ownership, and failure modes. If the account is still in implementation, keep that problem separate from onboarding completion and first value; our guide to setup completion versus time to value covers that earlier stage.

Measure the workflow at the right level

Once the job is clear, build a small measurement set around it. The exact definition depends on the product and customer process; avoid importing a universal “healthy” threshold.

  • End-to-end completion: the share of eligible instances of the target workflow that reach the agreed outcome in the product. State what makes an instance eligible and what counts as complete.
  • Coverage by account and role: whether the required people participate, rather than counting one frequent administrator as adoption for the whole account.
  • Time and rework: how long the workflow takes and where customers repeat entry, reconcile versions, or wait for a handoff. Use customer walkthroughs or a consented sample when this activity sits outside product telemetry.
  • Outcome quality: whether the completed work is accurate, timely, or useful for the customer’s stated goal. Usage alone cannot establish that result.
  • Workaround trend: whether the extra step is shrinking, stable, or growing after a specific change. Record the reason for the workaround so a raw count does not hide a necessary control or edge case.

Keep usage metrics, workflow completion, and business outcomes distinct. A higher event count after a product change does not by itself prove the customer’s process improved. Check whether the intended workflow became more complete and less burdensome, and ask the customer whether the change solved the problem.

What a weekly login can—and cannot—tell you

A regular login is a useful signal: the product has a place in someone’s routine. It is not proof that the product owns the important workflow, that every required role has adopted it, or that the customer has stopped doing critical work elsewhere.

For an established account, the better review question is: Which recurring customer job now runs reliably through the product—and what still depends on a workaround? Answer that with a workflow map, a small set of outcome-linked measures, and a direct customer walkthrough. Then you can tell the difference between a product gap worth fixing and a spreadsheet that is doing a job your product was never meant to do.

#product-adoption#saas-workflows#customer-success#revops#product-analytics

Rothenhall Partners

Be the company the AI recommends.

We own AEO, GTM, and RevOps as one accountable engine. See what that looks like for your company.