How far below average does each complaint sit?
Support and account-management complaints go with the largest negative gap of the eight types, and price complaints with the smallest. When a buyer criticizes a tool's support or account team, their rating sits about 1.6 points below the average other buyers gave the same product. Overselling is about −1.4. Usability, reporting, reliability, implementation and integration complaints range from about −0.7 to −1.0, and price is about −0.5. These figures describe how buyers rated products when they voiced each complaint. They do not establish that any one complaint caused the difference, and one buyer can mention several problems.
| The buyer says | Complaint | −1 0+1 | Praise |
|---|---|---|---|
| Support or the account team | −1.6 | +0.9 | |
| The tool was oversold | −1.4 | ||
| Hard to use | −1.0 | +0.7 | |
| Weak reporting | −0.9 | ||
| Bugs or outages | −0.9 | ||
| A painful implementation | −0.8 | ||
| Poor integrations | −0.7 | ||
| Price | −0.5 | +0.3 |
The complaint rows rest on dozens to hundreds of ratings each, with overselling and integrations among the smallest, so their place in the order is the least certain. Buyers can raise several complaints about one product, so the types overlap and the gaps are not independent effects. Each type was identified separately, so the kinds of comments captured may differ: read the order as a comparison of the complaint groups found, not a precise hierarchy of impact. Praise is shown only for the three areas where it was examined.
Read the bars against each other rather than one at a time. The finding is the difference in gaps across types: support, at −1.6, is more than three times the size of price, at −0.5. The same broad ordering appears against each buyer's other ratings, where most gaps widen a little, support and overselling stay at the top, and price stays at the bottom at about half a point. These are descriptive comparisons, not estimates of how much satisfaction would improve if one issue were fixed.
What the two measures mean
Both measures compare one rating with a baseline, and they answer different questions.
- Against the product's average
- The rating minus the average that other interviewed buyers gave the same product, leaving this rating out. It shows how far the rating that comes with a complaint sits from the product's usual score, which reduces baseline differences between products. It does not isolate the complaint's effect.
- Against the buyer's other ratings
- The rating minus the average of the other tools the same buyer rated in the same interview. It reduces the pull of general scoring habits, such as a buyer who scores everything harshly. Those other tools differ in what they do and in how satisfied buyers usually are with them, so this is a complementary benchmark and does not hold product quality constant.
- A complaint
- An identifiable negative statement, in the buyer's own words, about that aspect of the product they are rating. Mild criticism counts (a bit clunky, can be pricey); negations, hypotheticals and complaints about other products do not. One rating can fall under several complaint types, and each records what was said, not how severe it was.
Why do support complaints go with such low ratings?
The interviews do not establish why. Individual accounts show service experiences shaping a buyer's overall evaluation, and some buyers separate a capable product from their frustration with account management, response times or getting problems resolved. That is a plausible explanation of the pattern, not evidence that support problems lower every affected rating by the same amount.
The same split shows up elsewhere. A UX researcher at a large US retailer rated a user-research platform 4 out of 10 after praising the product in the same answer, saying its platform for usability testing “is incredibly intuitive and easy to figure out”, and then: “The only thing that sucks is their team; the people that work there are awful.” The praise for ease of use and the low score sit in the same answer.
Praise shows a smaller gap in the other direction. Buyers who praise a vendor's support rate its tool about 0.9 points above the product's average, against about 0.7 for praise of ease of use and about 0.3 for praise of price. In the three areas measured both ways, the negative gap is the larger one, though praise and complaints were gathered differently, so the comparison is approximate.
How do buyers rate software they say was oversold?
About 1.4 points below the product's average, and about 1.6 below the buyer's other ratings. It is one of the smallest groups on this page, so its exact place in the order is less certain. The comments describe a mismatch between the expectations set during the evaluation and the capabilities, costs or effort found afterward: features that needed extra services, development work nobody planned for, timelines longer than the sales team described.
The two complaint types with the largest gaps are both about the vendor rather than the software: what the people selling and serving the tool do, rather than how the tool works.
Do price complaints go with lower ratings?
Only slightly: price shows the smallest negative gap of the eight. A buyer who calls a tool expensive rates it about half a point below the product's average, and the gap is similar against their other ratings. It shrinks slightly when ratings that also carry another complaint type are left out. That does not mean price is unimportant in purchase, renewal or replacement decisions; buyers can be unhappy with a price and still rate the software relatively well.
Many price complaints come from buyers who rate the tool well. About four in ten of the ratings that carry one are 8 out of 10 or higher, against about one in eight of the ratings that carry a support complaint. An ecommerce executive at a home-furnishings retailer rated a project-management tool 9 out of 10 and put the whole gap to a perfect score down to cost: “The only thing I would say that wouldn't take them to a 10 is their pricing. They're fairly expensive.”
When buyers say a tool is expensive and in the same breath say it is worth it, the rating runs the other way, about a point above the product's average. A healthcare marketing leader put it in one sentence about a data provider: “You know, they're expensive, but they work, so I'm going to give them a nine.” These accounts show buyers acknowledging a high price while still judging the product's value positively; the interviews do not show how their rating would change at a different price.
We also looked for categories where buyers forgive a high price and categories where they do not. Email platforms are the only group large enough to read, with price complaints sitting about 0.6 points below the product's average. Outside email the groups are too small, and the two measures disagree too often, to separate one category from another, so this page does not rank them.
The counter-current: part of the gap is the scale
Negative gaps exceed positive ones in all three areas measured both ways. That could suggest buyers weigh problems more heavily than strengths, but the scale matters: buyers mostly rate their tools above the middle of a 1 to 10 scale, so there is more room for a rating to fall than to rise. These comparisons cannot separate the scale, which comments were captured and how they were worded, and genuine differences in how buyers weigh problems and strengths.
What this means for a software trial
These are our recommendations, drawn from the patterns and accounts above, not a process buyers describe. Support failures and unmet sales promises go with particularly low ratings, so they deserve deliberate testing alongside capabilities, integrations, reliability and cost. The order below is a list of issues worth investigating, not a proven ranking for every buyer.
- Support, before you sign. Where the trial allows, submit a realistic request and judge the quality of the answer, the expertise behind it, how it is escalated and whether it is resolved, not just the speed. Ask references about service after onboarding, including what happened when their account team changed, and clarify which support the contract includes.
- Every promise that decides the purchase. Write down the capabilities, integrations, performance commitments and timelines the sales team presented. Test the workflows that matter on realistic data, identify what needs extra modules, services, configuration or development, and confirm those dependencies, costs and responsibilities before you sign.
- Daily use, by the people who will do it. Have representative users test their actual workflows, not just watch demos built for purchasers or admins. Check navigation, task completion, error handling and the effort a task takes every time, over a testing period that fits the product's complexity.
- One report you rely on. Recreate an important report on representative data, checking definitions, accuracy, flexibility, exports and the effort involved. Reporting complaints go with ratings about 0.9 below the product's average.
- Stability and connections. Review the vendor's incident and SLA information, and ask how failures are communicated and resolved. Test the connections to the systems you depend on, including data accuracy, recovery from failures and ongoing maintenance.
- Price, alongside value and total cost. Price complaints show the smallest gap on this page, and many buyers who mention expense still rate the tool highly. That does not mean price deserves less weight. Compare licensing, implementation, support, usage charges, operating costs and expected benefits against your budget and requirements.
Want a review of your own stack?
Get my review →What not to read into this
That a complaint causes a low rating. The complaint and the rating come from the same conversation, often the same answer, and interviewers ask buyers who give a low score what is driving it. So low raters have more reason to list complaints, and some of each gap on this page may run from the rating to the complaint rather than the other way. What the page can compare is the size of the gap across complaint types, each computed the same way.
That price does not matter. It matters to what buyers do; the pages on why buyers keep software they rate poorly and when buyers start shopping show cost in both decisions. This page measures something narrower: how far a rating sits from its benchmarks when the buyer complains about price.
That one complaint type is more common than another. Each type was found with its own searches, so the counts behind the rows reflect how each area was searched and are not a measure of how often buyers raise it. The same process can also shape which examples enter each gap comparison, complaints co-occur, and the groups differ in size. The page shows which complaint types go with larger or smaller gaps, not their independent impact or how common they are.
That a vendor with a support complaint has bad support. These are individual buyers describing their own accounts, pooled across hundreds of products. No figure here describes any vendor, and none is named.
Common questions
Which software complaints are associated with the lowest ratings?
Complaints about the vendor's support. When a buyer criticizes a tool's support or account team, their rating sits about 1.6 points below the average other buyers gave the same product, and about 1.8 below the other tools the same buyer rated. Complaints that the tool was oversold come next, at about 1.4. Complaints about ease of use, reporting, bugs or outages, implementation and integrations sit between about 0.7 and 1 point below. Price complaints show the smallest gap, about half a point. These are associations, not the number of points each problem causes.
Do buyers who complain about price rate software lower?
Slightly. A buyer who calls a tool expensive rates it about half a point below that product's average, the smallest gap of the eight complaint types, and about the same against the buyer's other ratings. About four in ten of the ratings that carry a price complaint are still 8 or higher, and buyers who say a tool is expensive but worth it rate it about a point above the product's average. That does not make price unimportant in purchase or renewal decisions.
Why are support complaints linked to such low ratings?
The interviews do not establish why. Some buyers who complain about support separate it from the product: they call the platform capable and describe frustration with account management, response times or getting problems resolved. Praise shows a smaller gap in the other direction: buyers who praise a vendor's support rate the tool about 0.9 points above its average, more than buyers who praise ease of use or price.
Do complaints cause low software ratings?
This data cannot show that. The complaint and the rating come from the same conversation, and buyers who already rate a tool low are asked what is driving the score, so low raters have more reason to list complaints. What the data does show is how the gaps differ between complaint types, each computed the same way: the gap that goes with a support complaint is more than three times the size of the one that goes with a price complaint.
What should you test in a software trial?
Our recommendation, not a practice buyers describe, is to test deliberately for the weaknesses that go with the lowest ratings, alongside capabilities, integrations, reliability and cost. Where the trial allows, submit a realistic support request and judge how well it is resolved, not just how fast, and ask references about service after onboarding. Test the workflows and promises that decide the purchase on realistic data, have daily users run their real work, and recreate one important report. Weigh price alongside value and total cost.
For the rating at which buyers start looking at alternatives, see when buyers start shopping; for why low-rated tools stay anyway, why buyers keep software they rate poorly; for how the same product rates at different company sizes, ratings by company size; and for one category's complaints in detail, the top complaints about SMS platforms.
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.