Read Slido results on a forecast call by checking the answer settings, allowing for overlapping percentages, and clearing test votes.
Key takeaways
- Match your interpretation to the question and answer settings.
- Read each percentage as a share of respondents; those shares can overlap.
- Remove test responses before collecting meeting votes.
You ask the room where they think the quarter will land, the results appear on screen, and the conversation shifts from opinions to numbers. The trouble is that a poll chart looks more precise than it is. A bar that looks like a clear majority can mean something very different depending on how the poll was set up, who voted, and whether the numbers were checked before anyone saw them.
Revenue leaders who run polls on forecast calls can make a few mistakes. They read percentages as if they must add up to 100%. They assume every vote came from a serious answer. They let a test response sit in the results. None of these mistakes is hard to avoid, but each one can push a team toward the wrong commitment.
This post covers three rules for reading Slido results on a forecast call. You can apply them before the next poll goes live.
Why poll results mislead on forecast calls
Forecast calls are full of pressure. Someone wants a single number, the CFO wants confidence, and the sales director wants to show momentum. A poll gives everyone a shared picture quickly, which is valuable. It also gives the room a feeling of certainty that the data may not earn.
Three things make the picture unreliable. The first is the poll format. A question that allows several answers produces totals that do not behave like a question with one answer. The second is the audience. Not everyone who votes has the same view of the pipeline, and some people vote to be polite, to test the tool, or to follow the crowd. The third is the data itself. If someone tested the poll before the call and their response was never removed, the chart shows a vote that no one in the meeting cast.
The rules below cover settings, arithmetic, and test-vote cleanup. Start with the settings, then the math, then the clean-up. If you do them in that order, the numbers on screen will be much easier to trust.
Rule 1: check the poll settings before you read a percentage
The first rule is to look at how the poll was built before you interpret any result. This sounds obvious, but many people read a chart the moment it appears and never ask what question the chart answers.
Slido's own guidance for multiple choice polls points to the settings for this reason. The help article on creating and running a multiple choice poll tells the host to select the Poll settings in the upper right corner to see the available options. Make that a habit. Before the poll opens, and again when you read the results, confirm whether the poll allows one answer or several.
Why does this matter so much? Because a single-answer poll and a multiple-answer poll ask different things. In a single-answer poll, each person picks one outcome, and the percentages describe how the room split. In a multiple-answer poll, each person can pick several outcomes, and the percentages describe how often each outcome was picked. If you read a multiple-answer chart as if it were a single-answer split, you will draw the wrong conclusion about the room.
Consider a poll on which deals people expect to close this month. If the poll allowed several answers and a participant ticked three deals, each of those deals gained a vote from that person.
For forecast work, the practical rule is this. Before you say anything about the results, write down the question as the audience saw it and whether they could pick one answer or many. Then describe the result in those terms. "Most of the room picked the upside case" is a fair description of a single-answer poll. For either format, claim a majority only when the results show one, and describe the upside case as possible only if the question asked about possibility. Those sentences lead to different follow-up questions. For clearer scenario wording, review forecast categories.
Settings also tell you what the poll does not capture. A single-answer poll shows different choices across respondents, but cannot capture several selections from the same respondent. If the answer list is long, people may choose the first reasonable option instead of the true one. Reading the settings and the wording together gives you a clearer sense of what the chart can and cannot support.
A quick test helps here. Before the poll opens, ask yourself what you would do if the top option won with a clear lead. If the answer would not change your forecast, the poll is probably the wrong tool for that question. A poll is good at showing how a room is thinking. It is weak at settling a number that depends on contract terms, approval timing, or customer budgets that few people on the call can see.
Rule 2: expect totals that may exceed 100%
The second rule follows from the first. When a poll allows several answers, the percentages may exceed 100%, so allow for overlapping selections before the chart appears.
Slido's help article addresses the question directly, asking why results do not always add up to exactly 100%. The answer is that multiple answers change the arithmetic. The same article says that if multiple answers are enabled, there is a high chance that the sum of all results will exceed 100%.
The article also gives a useful extreme case. It explains that if every participant chose all of the available options, each option would display the full share, and the combined total would be that share multiplied by the number of options available. A poll with many options could therefore show a combined total far above 100%. You do not need that extreme to make the point. When everyone answers and some people select more than one option, the total moves past 100%, and the more options each person ticks, the further it climbs.
What should a revenue leader do with this? Three things.
First, do not add a multiple-answer poll's percentages together and treat the sum as a check on the forecast. The total need not equal 100%, so a total well above 100 percent does not mean the poll is broken, and a total of exactly 100 does not mean it is correct.
Second, do not compare a multiple-answer chart with a single-answer chart from an earlier call as if they were the same measure. If last quarter's poll asked people to pick one outcome and this quarter's poll allowed several, a jump in one option's percentage may reflect the format change, not a change in the room's view.
Third, change the question when you need a split. If you want to know how the room divides across scenarios, use a single-answer question. If you want to know which risks people see, a multiple-answer question is fine: report respondent counts or percentages of respondents, explaining that the shares can overlap. That wording keeps the number honest.
Illustrative example: overlapping selections
Adding those percentages gives 100% × 3. Each option's share is valid, but the combined percentage is not a split of respondents into separate groups.
Rule 3: clear test votes and keep the option list short
The third rule is about the data and the design. Before the call, run a test. After the test, reset the results so that your own vote does not appear in the live chart.
Slido's guidance covers this case. The help article notes that you can delete your testing response by clicking the three-dot menu next to your poll and selecting Reset results. It is easy to forget. You open the poll, pick an answer to see how it looks, close the test window, and start the call. Then the first poll shows a vote for an option that no one in the room chose. If the test vote was the only vote for an option that had no real support, the chart may show a small but misleading bar. On a close call, a single vote can change the story.
Make a checklist item out of this. Open each poll once, vote as a tester, confirm the display, and then reset the results before you share the poll with the room. If you are running several polls, reset each one, not just the first. Ask a colleague to check the results panel before the meeting starts.
Also keep the option list short. The same Slido article offers a clear principle here. It says that you can add as many options as you like, but less is more. It also advises that you be selective with your choices to help your audience engage more easily and keep the experience smooth for everyone.
For a forecast call, this means limiting the poll to the outcomes that matter for the decision. A poll with many revenue bands, each one close to the next, asks people to fine-tune guesses they may not be able to defend. Answers spread across neighboring bands can make the chart look more exact than the room's views really are. A shorter list with clear, distinct outcomes gives people a fair choice and gives you a result you can act on.
Short lists also help with the multiple-answer problem from Rule 2. When there are fewer options, each option's share is easier to interpret, and the room can see at a glance which outcomes drew votes and which did not.
Putting the rules into your call
These three rules are easiest to use as a routine, not a set of ideas to remember in the moment. Here is a practical sequence for a forecast call.
Before the call, decide what each poll is for. Write the question first, then the options. Ask whether the answer should be a single choice or several. If you want a split of expectations, choose one answer. If you want to know which concerns exist, choose several. Keep the list short. Write down, in one line, what the result will be used for, so that the number has a purpose before anyone sees it.
Next, test the poll. Vote once, look at how the chart displays, and then reset the results. Check the settings in the upper right corner of the poll so you know the format you are about to share. If a colleague is co-hosting, tell them which poll is live and what the expected answer type is.
During the call, open the poll and read the settings out loud or on screen before you describe the results. Say the question as the room saw it. Describe the chart in the right terms. If the poll allowed several answers, explain that selections can overlap. If the result surprised you, ask a follow-up question before you change the forecast. A poll should start a conversation about the number, not replace the work behind it.
After the poll, avoid writing the result as a single fact in the notes. Record the question, the format, the number of respondents if you can see it, and the result. Someone reading the notes next week should be able to tell whether a tall bar meant a majority picked the option or simply that many people ticked it.
What to do after the call
Polls are most useful when they connect to the rest of the forecast process. After the call, take the results into the same review you use for pipeline data, deal notes, and historical close rates. A poll can tell you what people believe at a moment in time. It cannot tell you whether a deal will close.
A useful habit is to write one follow-up question for each result that moved you, and assign it an owner. If the room leaned toward a later close date, the follow-up might be to check the contract stage of the deals that people named most often, and to confirm who is responsible for each one. The poll then becomes a list of things to verify, rather than a number that feels settled because everyone saw it on screen. Use a close date change log to keep that follow-up connected to the deal record.
Slido results can also be useful as a record. Cisco's help article on host reports notes that if you use Slido for Q&A or polling, you can find that information in Slido's report. If your team uses a report after the call, check that the report matches what you saw on screen, and note the poll format next to each result. A report without the format is easy to misread later.
Finally, close the loop with the people who answered. If the poll showed that the room expects a number your model does not support, say so and explain why. If the poll showed agreement you did not expect, ask the people who voted for that option what they know that the pipeline does not show. The poll is a prompt for better questions, and the best forecast calls use it that way.
A quick checklist for your next forecast call
Today, open the poll you plan to use next and write down its exact question, answer type, and intended decision before testing it.
Questions about poll results
Does allowing multiple answers guarantee a total above 100%?
No, enabling multiple answers permits overlapping selections but does not require anyone to choose several options. Read the actual results before describing their total.
Can a poll settle the forecast?
No, a poll records respondents' views rather than proving that a deal will close. Use a surprising result to identify a deal-level question to investigate.