How often do companies go back to software they left?
Rarely, in Alium’s interviews. Fewer than 2% of the departures buyers describe are associated with a return to the tool the company left. That describes the reported histories, not the probability that any given switch is reversed. Among the returns, about half are email, SMS and reviews software. The reason buyers give most often is that the replacement’s results disappointed; the next is that the team could not run it. And about a quarter of returns reverse a consolidation, most visibly SMS and email moved onto one platform and then split again.
Two terms carry the page. A departure is a tool a buyer says their company stopped using. A return is an identified reversal: the company went back to a tool it had stopped using for the same job, after running that job on something else in between, whether another vendor, an in-house build or a module of a platform it already had. Usually that tool belongs to a former vendor; in a few cases the company kept the vendor and went back only to its tool for that job. A tool that was only switched off and later switched on again, with nothing in its place, is not counted, and neither is a person bringing a tool they used at a previous employer to a new one. That is a different pattern, covered in do marketing tools follow people. About three returns in four had already happened at the time of the interview. The rest were decided and under way, or covered only part of the job; they are included, but a planned or partial return is not a completed reversal.
The under-2% figure is an observed share from interviews, not a complete return rate. A return is counted only when a buyer describes it, and an interview does not capture every decision. Some departures may be too recent for a return to have happened yet, and some older returns may go unmentioned because the person interviewed does not know or recall the full history. So the observed share may differ from how often returns eventually happen, and neither the direction nor the size of that difference can be established here. It is also not a vendor churn rate: the reference population is the departures buyers describe, not any vendor’s customer base.
Which software do companies go back to?
Email, SMS and reviews tools are disproportionately represented among the identified returns. They are about 18% of the departures buyers describe and 49% of the returns. Analytics, BI, testing and attribution tools are about 22% of departures and 10% of returns. That is a substantial difference in composition, and it suggests that former vendors in the first group reappear more often. With a few dozen returns and incomplete histories, the size of the gap cannot be pinned down precisely.
Each strip is the category distribution within its own set, and adds up to 100%. The first is every tool buyers describe having left; the second is every identified return. Both count each company and tool once, and counting departures once per company and vendor instead gives the same shares. Each case sits in the group of its main category, and the groups do not overlap. The return strip rests on a few dozen cases, so its proportions describe these returns; they are not return probabilities for each category.
The rest of the returns are spread thinly across commerce platforms and storefronts, site search, personalization, customer service, loyalty, social and a few others, with none large enough to say anything about on its own. Commerce is the one worth a sentence: several of its returns are companies going from a custom headless storefront or a custom checkout back to their commerce platform’s own. They never left the platform’s vendor, so these reverse an implementation choice rather than a vendor decision, and they sit at the edge of what counts as a return here.
One possible explanation, and it is ours rather than something buyers state: email, SMS and reviews tools produce results a team can compare quickly (revenue per send, conversions from a text, reviews collected per order), and several of the buyers who went back describe disappointing results from the replacement, which is consistent with that. The interviews do not establish that these results show up sooner across the market, or that faster feedback causes more returns. Analytics and BI migrations may differ in their criteria, their historical data and their dependencies, and fewer observed returns there does not show that those replacements succeeded more often. It could reflect switching costs, priorities, or which experiences buyers happened to discuss.
Why do companies go back?
Among the identified returns, disappointing results from the replacement are the most frequent main reason, followed by a mismatch between what the replacement needed and what the team could supply. Each return is assigned one principal reason from the buyer’s account, though several factors may have contributed. These are the reasons behind the observed returns, not an account of why replacements fail across the market.
| Why they went back | Share of returns | What buyers describe |
|---|---|---|
| The replacement’s results fell short | About 3 in 10 | Lower email or SMS performance, fewer reviews collected, weaker product recommendations, a platform that could not keep up with the business |
| The replacement needed more than the team had | About 1 in 4 | Every email needed coding, every search change needed a developer, dashboards non-technical staff could not build, specialist skills nobody could hire |
| Price | About 1 in 7 | The replacement got expensive, or the former vendor came back cheaper |
| The move failed partway | About 1 in 8 | An implementation that never launched, a migration that lost data, an in-house build that struggled |
| Something else | About 1 in 5 | The new vendor discontinued the product, a new decision-maker preferred the old tool, the old vendor improved, or no clear reason given |
Worse results is the commonest story, and in reviews it has the plainest number. A company that switched reviews platforms watched review collection fall: “How many ratings and reviews we were getting precipitously dropped,” the buyer says, “so we went back to Yotpo.” One brand that moved product recommendations into its email and SMS tools found that neither did the job as well, and went back to Rebuy.
The second story is fit. An ecommerce company had left Klaviyo for a more powerful, CDP-style email platform and could not find people to run it. The marketer who arrived afterwards puts it plainly: “We were probably never a good fit for that platform. It just requires more hands on deck.” The company went back to Klaviyo. Another brand had gone from Klaviyo to one email platform and then another that needed code for every send, and returned for the same reason: there were not enough developers. In site search, a business that had moved to a search engine where every change to the results needed a developer went back to the vendor whose tools let the business team make those changes without code.
Time on the replacement runs from a few months to about two years. One commerce buyer gave a new platform four months: “We lasted like four months and then decided to go back.” At the long end, a replacement email platform lasted about two years before the company broke the contract and returned to Salesforce Marketing Cloud.
Going back on a consolidation
About a quarter of identified returns reverse a consolidation. The company had moved the tool’s job onto a platform that already did other jobs (its email platform, its SMS platform, its commerce suite, its marketing automation), then went back to the tool it had used before. Email and SMS are the clearest case, and it runs in both directions. This describes what the returns are made of. It does not estimate how often consolidations fail, because the interviews offer no comparable count of consolidation attempts and their outcomes.
Some brands moved SMS into Klaviyo or another email platform and later moved it back out to Attentive. “We actually just joined Attentive again,” one says. “We left them previously and did Klaviyo for SMS.” Others went the opposite way: they moved email into their SMS platform, some after being sold on running both channels in one place, and later moved email back to Klaviyo. One of those was, at the time of the interview, “in the final week right now of transitioning back to Klaviyo with better support guaranteed, a dedicated rep.” Buyers in both directions describe the same problem: the added channel performed worse on a platform built around the other one.
The same shape turns up beyond the email and SMS swap. Reviews moved into an email platform and back out after the migration lost reviews; product recommendations moved into email and SMS tools and back to a specialist; personalization moved into a suite and back to Dynamic Yield; high-volume operational email moved into a marketing-automation platform and was being moved back to a dedicated sending service because it was jamming the platform’s other processes.
Each reversal has to be read against how often the move is attempted. Moving SMS into the email platform is a consolidation buyers commonly describe (see the great consolidation), so its reversals may be prominent partly because the move itself is common. Without a counted population of attempts and outcomes, the interviews cannot show that SMS consolidation is more failure-prone than any other. Single cases are told in more detail on SMS migration timelines and the top SMS complaints.
The counter-current: price pulls both ways
Price sends companies back in both directions. Some had left a cheaper tool for a more capable one and went back when the bill arrived: a contact-center tool far pricier than the helpdesk it replaced, an enterprise social suite that, in one buyer’s words, “priced itself out of our budget, and we had to go back to HubSpot.” Others left for something cheaper and paid again when it did not work: a free session-replay tool, a lower-cost testing tool that was too hard to launch tests in. Former vendors notice. One SMS buyer says Postscript “kind of came back and was asking if we would be willing to migrate back over”; an address-validation vendor promised to beat the incumbent’s price at renewal; a testing buyer recalls “some sort of discount rate when we went back.”
Want a review of your own stack?
Get my review →What this means before you leave a tool, or go back to one
These are our recommendations, drawn from the return stories above, not a process buyers describe. They point to three things worth settling before leaving an incumbent: how the replacement performs, what it takes to operate, and how it fits with the rest of the stack. Five checks:
- Record a baseline before you migrate. Write down the outgoing tool’s key outcomes (revenue per send, conversion from a text, reviews collected per order) along with the traffic, customer mix, season and metric definitions behind them, then compare the replacement on the same measures over a meaningful period. Some of the returns followed disappointing results; the right measures and the right period depend on the category.
- Check whether a rollback is practical. Before shutting the incumbent down, check whether its data, configuration, integrations and sending or operating capability can be kept long enough to validate the replacement. A staged migration or a temporary overlap may reduce the risk where it is feasible; weigh it against duplicate licenses, data syncing, privacy, security and the cost of running both. One company that moved email onto a combined email and data platform never closed its Klaviyo account, because it was month to month: “we just reduced our spend with them significantly,” the buyer says, and the account stayed active while sending no email. When the project was pulled, the old account was there to go back to.
- Check that the team can operate it. Identify who will run the everyday workflows, reporting, configuration, integrations and troubleshooting, and test the replacement with the people actually available, including any developers, consultants or agency and what they cost on an ongoing basis. About a quarter of the returns had a main reason involving skills or staffing the team could not supply.
- Evaluate each consolidated function on its own. When a platform takes on a job another tool did, test whether its module meets that job’s requirements, not just whether the platform does its main job well. Compare results, operating effort, the benefit of the integration and the cost of reversing the move.
- If you go back, renegotiate. Confirm the vendor’s current capabilities, implementation requirements, how your data will be restored, service commitments, migration help and pricing. Some returning buyers describe discounts or offers from the former vendor, but those terms are not guaranteed. Check that the vendor has addressed the reason you left, and whether going back beats the other options.
What not to read into this
It is not a measure of how often switches fail. A return is one way a switch can go wrong; a company that stays on a replacement it dislikes, or moves on to a third vendor, does not show up here. It is not evidence that consolidation is a mistake. Most departures in these interviews stay departed, and without a count of consolidation attempts and their outcomes, the reversals here cannot say how often consolidations work. It is not a verdict on any vendor. The platforms companies went back to are also platforms other companies left, and a vendor appearing as the one left in a case here says nothing about how its customers fare generally. And fewer analytics returns does not mean analytics switches go better. The data shows where returns happen, not why; the reading above is ours.
Common questions
How often do companies go back to software they left?
Rarely, in Alium’s buyer interviews: identified returns are fewer than 2% of the departures buyers describe. That is an observed share of reported histories, not a complete return rate. A return is counted only when a buyer describes it, some departures may be too recent for a return to have happened, and some older returns go unmentioned because the person interviewed does not know or recall the full history.
Which software are companies most likely to go back to?
In these interviews, email, SMS and reviews tools. They are under a fifth of the departures buyers describe and about half of the identified returns. Analytics, BI, testing and attribution tools are about a fifth of departures but only one return in ten. Those are shares of the returns observed, not return probabilities for each category. Klaviyo is the platform companies most often went back to for email, and Attentive for SMS.
Why do companies go back to a vendor they left?
Most often because the replacement’s results disappointed, the main reason in about three returns in ten: lower email or SMS performance, fewer reviews collected, weaker recommendations. In about a quarter, the replacement needed more skill or engineering than the team had. The rest went back over price, a move that failed partway, a product the new vendor discontinued, a new decision-maker, or another or unstated reason.
Do companies regret moving SMS into their email platform?
Some do. About one identified return in four reverses a consolidation, and the clearest pattern among those is email and SMS, in both directions: brands that moved SMS into their email platform and later moved it back out to a specialist, and brands that moved email into their SMS platform and later moved it back to Klaviyo. This does not show that such moves usually fail. Moving SMS into the email platform is commonly described, so its reversals may stand out partly because the move is common, and the interviews do not count consolidation attempts and their outcomes.
Should you go back to a vendor you left?
The interviews show companies doing it, not whether it was the best option. Our recommendation, not a practice buyers describe, is to compare the replacement with a baseline recorded from the old tool before the move, and to renegotiate if you go back: confirm current capabilities, data restoration and pricing, and check that the reason you left has been addressed. Some returning buyers describe discounts or offers from the former vendor, but those terms are not guaranteed.
For why email platforms get replaced in the first place, see why companies switch email platforms; for the pressure to run fewer tools, the great consolidation; and for tools that travel with people rather than staying with companies, do marketing tools follow people.
A tech stack review, powered by the buyers who've been there
What matters most to you, a shortlist for your stack, and the next steps to take — built from hundreds of verified buyer interviews.
Get my stack reviewFree · no password · your answers are anonymized before they inform research like this.